Doctrine of command responsibility

THE CONTEMPORARY LAW OF SUPERIOR RESPONSIBILITY

American Journal of International Law v.93, no. 3, July 1999 

By Ilias Bantekas⁽¹⁾

 

“x x x.

II. Legal Nature of the Doctrine

The Existence of a Duty 

People who fail to prevent or punish their subordinates' illegal acts are liable under the doctrine of command responsibility. Accordingly, the doctrine does not address actions but omissions. Most systems of criminal law punish unlawful positive conduct, thus requiring the commission of an act. Exceptionally, common law jurisdictions recognize personal liability for omissions only where a specific duty is prescribed either by statute or by the common law.⁽²²⁾ Such duties include those arising from familial,⁽²³⁾ contractual,⁽²⁴⁾ doctor-patient⁽²⁵⁾ and voluntary assumption of care⁽²⁶⁾ relationships, as well as the duty to avert injurious consequences arising from one's own fault.⁽²⁷⁾ 

Express formulations of duties pertaining to commanders are found in Article 86 of Geneva Protocol I (1977), Article 6 of the International Law Commission's (ILC) Draft Code of Crimes Against the Peace and Security of Mankind,⁽²⁸⁾ Article 7(3) of the ICTY Statute,⁽²⁹⁾ Article 6(3) of the ICTR⁽³⁰⁾ Statute and Article 28(2) of the ICC statute.⁽³¹⁾ All these instruments give commanders the responsibility to prevent their subordinates from violating the laws of war and the responsibility to punish their subordinates if the laws of war are violated. 

X x x.

 

III. Subordination

Establishing a Superior-Subordinate Relationship 

The required three elements of the crime have developed over time. Depending on the source from which authority is acquired, the existence of a superior-subordinate relationship can be established in two independent ways, de jure and de facto. The most obvious means for the assumption of power is through official delegation of command from a pertinent office. This authority evinces de jure command over other persons. However, a position of command cannot be determined by reference to formal status alone.⁽⁴⁷⁾ In the absence of a formal grant of authority, the accused must actually be found to possess the right to control subordinates.⁽⁴⁸⁾ Mere authority to control actions of others does not preclude a finding of command responsibility.⁽⁴⁹⁾ 

Such de jure command may be determined by reference to the position of the accused "in the overall organization, with a view to determining [his/her] institutional functions."⁽⁵⁰⁾ Primarily, therefore, the nature of the "organization" must be discerned, and specifically its capacity to delegate further authority. Formal executive structures, such as state entities, usually vest such authority by passing legislative acts, which provide evidence of de jure command. For example, the Akayesu case involved allegations of incitement and complicity of a Rwandan local civilian official for the actions of others resulting in genocide against the Tutsi population residing in his commune. The ICTR Trial Chamber adduced that according to Rwandese law, Akayesu's position as burgomaster placed him as (1) the head of the communal administration; (2) the officier de l'état, and (3) the person responsible for maintaining and restoring the peace.⁽⁵¹⁾ That was sufficient to establish Akayesu's de jure authority as a necessary element of his conviction for the crime of genocide. 

De jure power determines one's competence and jurisdiction in the sense that authority to take action or intervention is confined to a pre-defined field beyond which there exists no competency and subsequently no liability.⁽⁵²⁾ Nonetheless, de jure command is not the best determinant of actual authority, since problems of identification will arise when legislation is absent, obscure, or, even if available, it inadequately describes one's actual functions and the amount of authority actually exercised.⁽⁵³⁾ 
 
The Sources of de Jure Command 

Both the civilian and military components of every state's machinery constitute integral national defense structures. This hierarchical model is built upon a vertical scale, which seeks to effectively filter the dictates of the decision-makers down to the soldier on the battlefield. This vertical scale assumes four general stages of command. The first is policy command. That involves the power to determine policy objectives and, consequently, the power to commit or withdraw a state's armed forces. This is exercised by state leaders either personally or collectively.⁽⁵⁴⁾ The second is the phase of strategic command. The highest military authorities are responsible to produce a viable military plan to achieve policy command objectives. These plans are usually drawn up by the Joint Chiefs of Staff in conjunction with other senior government officials, and require authorization by those with policy command authority before implementation.⁽⁵⁵⁾ 

When the plan is finally authorized, it is passed down the chain of command for implementation on the battlefield. This is undertaken at the third operational level by senior military officers who command midlevel groupings of forces, mainly in the form of corps or divisions. Senior military officers do not issue orders directly to troops, but direct the commanders of smaller groupings.⁽⁵⁶⁾ At the end of the scale, tactical commanders exercise direct command over troops. At this level, persons of any rank may assume tactical command, since even a senior private can head a unit whose other superiors have been incapacitated from battle.⁽⁵⁷⁾ 

This structure is divided differently in cooperative international military actions. In the recent United Nations command structure,⁽⁵⁸⁾ for example, authority for both policy and strategic command is vested in and exercised by the Security Council and, to a very limited degree, by the Secretary-General.⁽⁵⁹⁾ A UN operation consists of a Force Commander who exercises operational command over contingents provided by member states.⁽⁶⁰⁾ For the duration and purposes of the specific operation military personnel under the Force Commander are "international personnel," although they remain in their national service.⁽⁶¹⁾ In practice, Force Commanders have been granted "full authority" over their forces and have as a result been held "operationally responsible" for their performance.⁽⁶²⁾ 
 
De Facto Command and the Concept of Control 

If liability for the acts of others is established through the element of authority over subordinates, it follows that the mere existence of such authority, whether acquired de jure or de facto, renders one a superior for the purposes of Article 7(3) of the ICTY Statute.⁽⁶³⁾ While instances of de facto command authority are a phenomena of civil warfare, their existence was contemplated even in the context of inter-state conflicts. The United States Military Tribunal in the Sadaiche case pointed out that "superior means superior in capacity and power to force a certain act. It does not mean superiority only in rank [since] it could easily happen in an illegal enterprise that the captain guides the major, in which case the captain could not be heard to plead superior orders."⁽⁶⁴⁾ In that case, the accused, a commanding officer of a POW camp, was held liable because he was led to acquiescence by his "more powerful adjutant."⁽⁶⁵⁾ Power to force a certain act inevitably involves the power to demand and an actual capacity to impose obeisance. Therefore, usurpation of authority contrary to national law would not prevent an individual from being convicted on the basis of command responsibility for international crimes committed if the person had de facto authority. 

This view was upheld during the preparatory conferences that led to Protocol I (1977), where reference to "commanders" in Article 87 encompassed persons in command "at the highest level to leaders with only a few men under their command."⁽⁶⁶⁾ This consistent irrelevance of rank in attributing superior responsibility indicates that the international law-making institutions look to actual and effective control,⁽⁶⁷⁾ rather than formality.⁽⁶⁸⁾ Article 87 of Geneva Protocol I (1977) extends the legal obligations of commanders beyond troops under their command to cover in addition "other persons under their control."⁽⁶⁹⁾ This obligation is applicable to superiors at all levels of command.⁽⁷⁰⁾ If the case were different, superiors with ample means to intervene in crimes committed by troops under their control, but not under their command, would be fully justified in being passive. Accordingly, the concept of "command" is not the only operative term for ascribing command liability, as the text of Article 87 extends the obligations of commanders to troops under their control.⁽⁷¹⁾ 

Evidence of de facto control requires proof of a superior-subordinate relationship.⁽⁷²⁾ The elebii judgment recognized that the accused, Delali, by special authorization of his local War Presidency, was authorized to negotiate and conclude important contracts and agreements on their behalf, but noted that he never acquired any status which placed him in a hierarchy of authority creating a superior and subordinate relationship.⁽⁷³⁾ His function was described as one of "co-ordination." This consisted of negotiating agreements for the President. That rendered him influential, but it did not create a superior-subordinate relationship.⁽⁷⁴⁾ Control in this sense must be effective. However, partial control will suffice when superiors have not exercised their potential for full control, in an attempt to evade personal liability.⁽⁷⁵⁾ Similarly, when troops not normally under one's command are assigned to another's forces, they too are considered to be forces under the special commander.⁽⁷⁶⁾ 

A superior-subordinate relationship finally requires a chain of command.⁽⁷⁷⁾ Thus, tactical commanders exercise "direct subordination" over troops assigned to them, while commanders of occupied territory (executive commanders) enjoy "indirect subordination" over a civilian mob which they are under a duty to prevent from committing crimes against other civilians in the territory occupied.⁽⁷⁸⁾ Since a chain of command is a prerequisite for the exercise of superior authority, it follows that one cannot be termed a superior without corresponding subordinates.⁽⁷⁹⁾ That is why staff officers, who, irrespective of their rank, do not command troops, are only responsible when their participation in the delivery and execution of criminal orders is proven.⁽⁸⁰⁾ The sole exception to the requirement of subordination is the case of executive commanders in occupied territory.⁽⁸¹⁾ Their responsibility is coextensive with their area of command,⁽⁸²⁾ and is not dependent on the persons they command or the exercise of effective control. They are responsible for the behavior of the occupation forces as well as the civilian population in the occupied territory.⁽⁸³⁾ 
 

X x x.

 

V. Types of Command And Extent of Liability

Superior authority necessitates a chain of command.⁽¹²⁰⁾ While operational command supposes vested authority over units linked to the operational superior through the leader of each unit, tactical command refers to a state of actual control over a defined number of subordinates. Command and control are thus not the same thing but are evidently inter-related.⁽¹²¹⁾ The following is an analysis of the necessary and reasonable measures expected of operational, tactical and POW camp commanders, as well as the extent of their liability in accordance with either the people they command or control, or the territory they occupy. 
 
Operational Commanders 

The Yamashita case is a fine example of operational command. The accused, Supreme Military Commander of the Japanese Imperial Army on the Philippines and Governor of the islands, was charged with breaching his duty to control his troops, thereby permitting them to perpetrate numerous and large-scale offenses.⁽¹²²⁾ Despite his de jure command, Yamashita, who assumed his post only eleven days before the invasion by United States forces,⁽¹²³⁾ ordered a partial evacuation. He then split the remainder of the Japanese forces into three separate fighting units, ceding full command of two, and commanding the other himself.⁽¹²⁴⁾ Contrary to his orders, the evacuation did not occur, and Yamashita was left isolated in a remote mountainous region, apparently lacking communications with his headquarters and the two other commanders.⁽¹²⁵⁾ 

Yamashita's conviction was upheld for acts of troops beyond his de facto control, on the ground that operational command responsibility cannot be ceded for the purposes of the doctrine of command responsibility even though the specific aspects of such command are actually ceded to others.⁽¹²⁶⁾ This rule, referred to as the "delegation principle," is recognized as a general principle of criminal law.⁽¹²⁷⁾ It strains the mind to consider the possibility of upholding criminal responsibility in cases where both de facto control is missing and de jure command was already ceded for military purposes and not for the purpose of escaping criminal responsibility.⁽¹²⁸⁾ Certainly, it must have seemed to the Supreme Court that authority was delegated for the purpose of evading responsibility which raised the risks of crime.⁽¹²⁹⁾ Command responsibility cannot be avoided. 

The test followed in subsequent cases strikes a reasonable balance between knowledge of subordinate offenses and material capacity to act. Thus, the murderous onslaught of Japanese forces on the Chinese city of Nanking under the command of General Matsui, rendered the General liable even though he issued instructions to abide by the laws of war, albeit lukewarm instructions at best, because he had every opportunity and ability to act to prevent and stop the crimes.⁽¹³⁰⁾ It is now well established that operational commanders must exercise the full potential of their authority to avert war crimes. They will not be exonerated in cases of non-assertive orders or failure to supervise their implementation.⁽¹³¹⁾ Similarly, the United States Military Tribunal in the High Command case demanded a "personal dereliction," for the application of the failure to act doctrine,⁽¹³²⁾ indicating that command responsibility was not viewed as a form of vicarious or strict liability. Personal dereliction in this sense is tantamount to a failure to supervise. That supervision is commensurate with the exercise of actual control in the given circumstances. Knowledge of every aspect of operational command is not required.⁽¹³³⁾ Rather, commanders may lawfully, within reason, presume that "details entrusted to subordinates will be legally executed."⁽¹³⁴⁾ 
 
Executive Commanders 

Executive or occupation commanders are those individuals vested by the occupying state with supreme governing authority in the occupied territory. The difference between operational and executive commanders is that while the former are responsible for the acts of people under their command or control, the latter are accountable within the territory they occupy to assure that the rights of civilians and POWs therein are fully protected. In the case of executive commanders, subordination is unimportant, their responsibility is co-extensive with their appointed command structure.⁽¹³⁵⁾ This responsibility cannot be limited, even by a commander's own state's legislation.⁽¹³⁶⁾ Since executive superiors are not bound under the rule of subordination, they have a duty to prevent and punish crimes which involve auxiliary or allied forces stationed within their area of command.⁽¹³⁷⁾ A similar duty exists in cases of inter-communal rivalry.⁽¹³⁸⁾ Commanders are also obliged to enforce discipline in related events in order to protect both local majorities and minorities. 

In cases where executive commanders also perform operational tasks, they must nonetheless be aware of any occurrences within their territory, even if they are not present.⁽¹³⁹⁾ Hence, the exercise also of operational command does not absolve them from their executive command obligations. Acquisition of information can be achieved by demanding constant and reliable appraisal of all pertinent facts.⁽¹⁴⁰⁾ This generally involves establishing an effective reporting system which would ensure delivery of adequate and up to date information to the commander.⁽¹⁴¹⁾ The duty of executive commanders to protect and care for the welfare of civilians is a strict responsibility element in the actus reus definition of their command responsibility.⁽¹⁴²⁾ 
 
Persons Entrusted with the Care of Prisoners 

According to a basic principle of military law and practice, once under captivity, prisoners become subordinates to the captor commanders,⁽¹⁴³⁾ and are deemed to be in the hands of the captor state.⁽¹⁴⁴⁾ Since Hague Convention IV (1907) POWs are customarily offered humane treatment.⁽¹⁴⁵⁾ Therefore, responsibility for their treatment rests with government officials, operational officers in command of formations who are holding the prisoners, as well as officials in departments concerned with prisoner welfare and any other civilian or military official being in direct and immediate control of prisoners.⁽¹⁴⁶⁾ It seems unreasonable, however, to hold government officials responsible without requiring proof of knowledge and a capacity to act.⁽¹⁴⁷⁾ However, rules that hold the Head of government and minister in charge of prisoner affairs responsible are reasonable, since these officials' duties are to acquire knowledge and act accordingly.⁽¹⁴⁸⁾ 

While the liability of POW camp commanders is based on direct subordination,⁽¹⁴⁹⁾ the responsibility of departmental officials charged with prisoner affairs can be justified only if assimilated to that of staff officers who exercise neither command nor control authority. They would incur responsibility only when there is proof of direct participation.⁽¹⁵⁰⁾ Since camp commanders are entrusted with the care of prisoners, subordination of any kind is irrelevant; their authority and responsibility extends over the entire institution and all its personnel.⁽¹⁵¹⁾ 
 

X x x.

 

VII. The Duty to Act

Article 7(3) of the ICTY Statute, Article 6(3) of the ICTR Statute, Article 86(2) of Geneva Protocol I and Article 28(1)(a) of the ICC statute firmly establish the existence of a duty to prevent and a duty to punish the crimes of subordinate persons.⁽¹⁹⁰⁾ They constitute distinct and independent legal obligations,⁽¹⁹¹⁾ and this is further affirmed by the use of the disjunctive "or" in Article 86(2) of Geneva Protocol I. These articles necessarily suggest a further "duty to control." 
 
The Duty to Prevent 

Article 86(2) of Geneva Protocol I and Article 7(3) of the ICTY Statute expressly establish a duty to prevent at the moment subordinates "are going (or "are about") to commit a crime."⁽¹⁹²⁾ The duty to prevent commences upon the preparation or planning of an offense by subordinates. Commanders are even responsible for failure to prevent if they fail to take into account factors such as the age, training or similar elements that point to obvious conclusions regarding the likelihood that such crimes would be committed.⁽¹⁹³⁾ The abundance of information clearly indicating a massacre at two Palestinian refugee camps in Lebanon that was available to the Israeli Chief-of-Staff and the Defense Minister rendered them "indirectly responsible" under a test of anticipated foreseeability lacking direct intent.⁽¹⁹⁴⁾ Their duty, according to a subsequent investigative report of a governmental commission, was to recognize the potential danger and to prevent the killings by doing everything in their power to stop them.⁽¹⁹⁵⁾ 

Superiors successfully discharge their duty to prevent subordinate crimes when they employ every means in their power to do so.⁽¹⁹⁶⁾ This also involves pointing out objections to a criminal plan, even if the accused has no further authority to intervene.⁽¹⁹⁷⁾ Reference to a duty to prevent has been limited thus far in the ICTY and ICTR.⁽¹⁹⁸⁾ In the Akayesu case, Trial Chamber I pointed out that whether a superior had the power to prevent the crimes of subordinates is irrelevant if the superior did not attempt to do so.⁽¹⁹⁹⁾ However, failure to condemn incendiary statements uttered at an already formed gathering would not be tantamount to approval.⁽²⁰⁰⁾ This conclusion should be approached with some caution since it arose in the unique circumstances of the Rwandan massacres. It may not reflect a statement of general law,⁽²⁰¹⁾ but may pertain to civilian superiors whose authority to intervene is unclear.⁽²⁰²⁾ 
 
The Duty to Punish 

A superior's "duty to punish" arises after the commission of an offense. It is predicated upon offenses by others which have already occurred, not future offenses.⁽²⁰³⁾ Punishment is, therefore, intended to deter the commission of future offenses. This duty to punish is distinct from a commander's preventive duty. For that reason, failure to prevent recurrence of future crimes is another aspect of the "duty to punish."⁽²⁰⁴⁾ Failure to punish, as a form of command responsibility, is a species of imputed responsibility for an omission and is neither akin to the common law concepts of "accessory after the fact"⁽²⁰⁵⁾ nor "misprision of felony."⁽²⁰⁶⁾ The duty to punish does not require a pre-existing relationship to those who perpetrated the offenses, as this would have been part of the incumbent superior's preventive duty at the time the offenses occurred. Thus even persons who assume command after such offenses have taken place are under a duty to investigate and punish the offenders. 

Since it will not always be feasible to initiate judicial proceedings, especially in the midst of military operations, the only available option may be to commence adequate investigations and report the findings to superiors. Tolerating criminal conduct, as evidenced by the failure to punish, is tantamount to acquiescence.⁽²⁰⁷⁾ When superiors are physically unable to prosecute, arrest, incarcerate or in any way discipline a subordinate, they must try to discover whether crimes were committed and, if they are ongoing, to find ways to stop them.⁽²⁰⁸⁾ If a crime is discovered, they must refer the case to the appropriate judicial or prosecutorial authorities, or to any other competent

authority.⁽²⁰⁹⁾ 

In conclusion, a superior should be held responsible for failing to take all measures that are "within his material possibility."⁽²¹⁰⁾ Trial Chamber I in the elebii case correctly disagreed with the ILC's demand that both legal and material possibility to take appropriate measures be present. The Chamber took the position that only the latter is required.⁽²¹¹⁾ It would be absurd for superiors to claim that they were not legally entitled to prevent an act of genocide when they were materially capable of doing so. 
 
x x x.

VIII. Conclusion

The interplay between humanitarian and criminal law at the international level suggests that the best approach to the doctrine of command responsibility should be through the concept of control. Assessment of de jure command must not mislead tribunals as to the accused's actual authority. This in no way widens the ambit of command responsibility, since it follows the established principle of international law that actual power is linked to responsibility. The interest of the state in the protection of its officials should not obstruct the development of these norms. If the international community seriously intends to combat gross human rights violations it has to do so irrespective of particular states interests in order to attain the interests of the broader international society. An ideal ground would be to strengthen the international criminal justice system in ways that would not jeopardize national security interests by applying the law uniformly regardless of the state or region involved. 

Increasingly, the need for review or re-assessment of humanitarian norms will arise, especially when cases come before tribunals, such as the ICTY and ICTR. Then, factors like the unpredictable nature of civil warfare will demand adaptation of established rules to fit these new circumstances. Since Yamashita, the law of command responsibility has moved to new realms. The question now seems to center around the issue of subordination and control, while fierce debate exists on the applicable mens rea standard. A presumption of knowledge in cases of notorious and widespread criminality is an emerging rule of customary law. The traditional "failure to act" doctrine is too limited to cover situations in which a superior's negligent or deliberate inactivity to control his or her troops causes crimes to be committed by them when under another superior's command. 

Furthermore, command responsibility should not be overlooked when examining the nature of a mixed conflict. The ICJ in the Nicaragua case determined that when an intervening state gives "direct and critical combat support" to an insurgent force and insurgent operations reflect "strategy and tactics wholly devised" by the intervening state, then an agency relationship is established.⁽²²²⁾ Indeed, despite the inconsistent understanding and application of the Nicaragua test by the various ICTY Chambers,⁽²²³⁾ in cases where an agency relationship is found to exist it is not far-fetched to consider the military or political leaders of the intervening state to be individually responsible for certain acts of their agent "subordinates." 

The obvious strategy for any prosecution for a breach of humanitarian law is to choose one that provides evidence of the accused's direct participation in the crime, such as ordering or inciting the actions, preserving command liability as a second option. In any event, however, prosecutorial authorities should not hesitate to indict persons who, while in positions of authority and with knowledge of subordinate criminality, intentionally or negligently fail to prevent or punish those persons who directly participated in the crime. 

X x x.”