The issue of the attribution of responsibility to states has emerged as a paradigmatic example of the schism in public international law . Various international tribunals have proposed their own tests in order to address this issue of attribution of the conduct of private individuals to state responsibility. In the firmament of this emerging jurisprudence, there is no North Star that points towards a universal and objective test that can be applied for the imputation of responsibility to a state.
Central to the question of attribution of this responsibility is ‘control’. The attribution doctrine is premised on the idea of control, which sets out the legal relationship between individuals and states. The rules of attribution are to be found in the International Law Commission’s Articles on State Responsibility, viz. Articles 4 to 11 . This attribution operates normatively, which is to say that it relates to the application of rules that determine whether there a sufficiently close association between the conduct in question and a state, so as to treat it as an act of the state. As Crawford notes, “the attribution of conduct to the State as a subject of international law is based on criteria determined by international law and not on the mere recognition of a link of factual causality.”
The acts of a private organization or group of individuals in one state is attributable to another state, by treating the group as an organ of the latter state, in which capacity all the actions of the group can be pointed to that state. Alternately, it may be established that the said group was acting under the instructions or control of the state, which would make certain acts as instructed by the controlling state to be attributable to it. Both of the above situations are essentially questions of state control, where the state has control the body acting as its ‘organ’ or acting under its instructions.
But there is a third situation for which there needs to be accountability. What if the group acting in one state, exceeds the orders of the state, and indulges in acts which were not ordered by the controlling state, thereby exceeding their instructions? Would such actions still be attributed to the controlling state? While there is no consensus that emerges from the various control tests, the specific requirements as laid down in various decisions are discussed and set out further in this discussion.
THE CONTROL TESTS
NICARAGUA and THE EFFECTIVE CONTROL TEST
In the leading case of Military and Paramilitary Activities in and against Nicaragua , the ICJ had laid down two tests for ascertaining state control – the ‘strict control’ test and the ‘effective control’ test. These tests were also subsequently confirmed by the ICJ in the Armed Activities on the Territory of Congo (Armed Activities) and Bosnian Genocide cases.
In 1984, the United States was charged by the Republic of Nicaragua with violations of customary law and treaty obligations by virtue of its involvement in military and paramilitary activities against Nicaragua. The United States was accused by Nicaragua of attacking its oil pipelines, ports and storage facilities. Additionally, there were complaints of violation of Nicaraguan air space and mining of their ports, and financing, arming and training counter-revolutionary forces (called contras) for overthrowing the government in Nicaragua in a civil war, which resulted in violations of International Humanitarian Law by these contras.
To assess whether violations of International Humanitarian Law took place during the civil war, the court adjudicated on the actions of three different units: those of the agents of the government of United States, those of the UCLAs (Unilaterally Controlled Latino Assets) and those of the contras.
The court held that while the acts of the first two groups can be attributed to the United States, the acts of the contras couldn’t be. To reach this conclusion, the court identified two different levels of controls that groups may be subject to: ‘strict control’ and ‘effective control’.
At the preliminary level, the court tried to determine whether the entity is under such a strict control of the interfering state, that the authorities of that entity can be treated as a de facto organ of the interfering state. ‘Strict control’ is based on total dependence, in which the organ must be shown to have no independence or autonomy and must have acted merely as an instrument of the outside power.
On the other hand, the ‘effective control’ test is based on partial dependence where certain acts of the entity are controlled or ordered by the intervening state. The ICJ does not use this test to determine whether an entity is the de facto organ of a state or not. In order to meet the effective control test, it must be demonstrated that, (1) a link exists by virtue of factors like arming, training and financing of the entity, and (2) a level of control that clearly indicates that the acts where ordered or executed under the instructions of the controlling state. This test is different from the strict control test to the extent that while the latter relates to the conduct as a whole of the entity, the former attributes responsibility only for specific conduct .
The traditional view continues to be that the effective control test is the proper standard for ascribing private conduct to a state and for allocating responsibility, unless primary norms or lex specialis dictate otherwise.
TADIC and THE OVERALL CONTROL TEST
The widest schism between the principle of state responsibility comes from test propounded by the ICTY Appeal Chamber in Prosecutor v. Tadić , known as the ‘overall control test’.
ICTY has the jurisdiction to prosecute persons who commit or order to be committed ‘grave breaches’ of the Geneva Conventions of 1949 . For that to be proved, the grave breaches must have been of an international character. The ICTY had to decide if the accused could be imputed for violations of the Geneva Conventions in the armed conflict in Bosnia and Herzegovina. This depended on whether the acts of the forces of the Republica Srpska could be attributed to an interfering state, in this case the Federal Republic of Yugoslavia (FRY), which would render the whole conflict international in its character.
While the Trial Chamber, using the ‘effective control’ test reached the conclusion that FRY could not be held responsible for the acts of the Republica Srpska, the Appeals Chamber overturned this decision. Both the chambers had based their decisions on the “general international rules of State responsibly which set out the legal criteria for attributing to a State acts performed by individuals not having the formal status of State officials” While this was a question of criminal liability, the Appeals Chamber had framed the question as one of State responsibility, namely, that of FRY. The tribunal ruled that the conduct of the Republica Srpska could be ascribed to the Republic of Yugoslavia, given that ‘as a whole’ these forces were under the ‘overall control’ of the Republic, thereby rejecting the effective control test.
As per the Appeals Chamber, the basic requirement in international law for the attribution of the actions of private individuals to state is that the state must exercise some degree of control over those individuals//groups . As regards the degree of control, the Chamber noted that it may vary according to the factual matrix of the case . This is also the position taken in the ILC’s Articles on State Responsibility which do not suggest any specific degree of control.
The court was of the opinion that the effective control test may be useful in ascribing responsibility with regard to ‘private individuals’ or ‘unorganized groups’, by establishing that there were specific instructions issued by the controlling state, or that the acts of the entity were later confirmed or approved by the state.
In cases of military and paramilitary groups i.e. organized and hierarchically structured groups, the proper test to apply is the ‘overall control test’. By proving that the controlling state exercises authority over the entity not only through arming, training and financing, but also helping them with coordination and general planning of the activity.
However, the ‘overall control’ test is not supported by the ICJ. In its decision in Armed Activities and Bosnian Genocide, the Court reaffirmed the ‘effective control’ test.
ANALYSIS OF THE TESTS
1. EFFECTIVE CONTROL TEST
The effective control test laid out in Nicaragua, seems to be the most popular test of all. But the persuasiveness of the test depends on whether it is premised on customary international law (depending on case law, state practice and opinion juris) or on the general principles of state responsibility, or even the general principles of international law.
But the decision in Nicaragua does not clarify or set out the source of this test. This is one of the primary criticisms of the test- that it does not seem to be supported by any principle of customary international law or even the general principles of international law. The threshold of control that is propounded in the case is not based on any state or judicial practice. Further, it was pointed out in the Tadić decision that the relevant judicial practice on effective control only points towards the cases where single private individuals act on behalf of the state, and there is nothing to show that the test has been applied in the context of armed groups or forces.
Later, in its re-affirmation of the ‘effective control’ test in Bosnian Genocide, the ICJ noted that the test coincides significantly with Article 8 of the ILC Articles on State Responsibility, which reflect customary law according to the Court.
Regarding the similarity between the Article 8 of the ILC Articles on State Responsibility, Cassese notes that this claim is entirely misplaced. Article 8 reads as follows:
“Article 8. Conduct directed or controlled by a State: The conduct of a person or group of persons shall
be considered an act of a State under international law if the person or group of persons is in fact acting on the instructions of, or under the direction or control of, that State in carrying out the conduct.”
Cassese notes that the conditions laid down in the test (instructions, directions, control) are not cumulative standards. Rather, as the article itself states, they are disjunctive. Hence, the view of the court that the articles coincide does not hold any water.
Further, it is submitted that even if the above claim is accepted at face value, the argument of the court falls into the circular reasoning fallacy. In the Draft Articles on State Responsibility, the ILC has relied on the decision in Nicaragua to devise the Article. It would appear that the court is using a corollary of the test itself to confirm its validity, which is clearly fallacious. Hence, the court’s reasoning in predicating the test on the authority of the Articles is clearly fallacious owing to this circularity in its reasoning.
Apart from not being based on any judicial or state practice, there is another objection to the threshold of the effective control test. The states must answer for the acts of those acting on their behalf, even if those individuals exceed the mandate or their scope of operation. Otherwise, the states can evade liability when they act through such entities that end up doing damage not foreseen by the state. It is a fundamental principle of agency that when a persons acting on behalf of another exceeds his authority, the person commanding must take responsibility for it. Needless to say, under the effective control test there is no method to prove that those unforeseen acts were ordered or executed under the control of the state. In such a case, the worst atrocities can go unaccounted.
As such, the Tadić test seems to be a more appropriate test to address such situations. The factum of support, financing and training itself shows a strong connection between the state and the organized group acting on its behalf. It is only logical for the state in such cases to bear responsibility for prohibited acts committed by those under its control or following its orders.
Another criticism of the effective control test stems from the ‘slippage problem’. “Slippage refers to the decline of government control or oversight, whether through regulation, ownership or delegation of functions traditionally associated with the state.” For example, Private Military Contractors are becoming common and provide armed combat and security services, and are more often than not, not integrated into national militaries and so they are not generally treated as organs of a state or as exercising governmental authority. In cases a PMC is used to launch a combative attack, its actions may be attributed by using the effective control test under Article 8. But when the PMC is used to provide security services, and the agent uses lethal force in the process, the effective control test may or may not apply in this situation. This slippage, resulting from the delegation of functions traditionally performed by the state to other entities, becomes even more apparent when the ultra vires acts of such entities do not attract Article 8.
As an alternate technique to address this accountability gap, commentators have suggested lowering the threshold of control required for such attribution. Using the overall control test and placing reliance on Article 7 of the Articles on State Responsibility , which deals with situations of excess of authority or contravention of instructions of the state, situations like these can be addressed.
...(download the rest of the essay above)