During the second half of the nineteenth century, numerous writers began exploring the issue of the liability of the state for failing to prevent the wrongful acts of private persons committed within its territory, connecting it with the issue of state liability due to the fault (Wiesse, 1898; [6]). Prior to this period, rare writers had examined this question. In the 18th century Hugo Grotius noted that a community’s responsibility for an individual’s actions depended on whether there was an unlawful conduct of the community itself. Accordingly, he stated that a civil community is not liable by the acts of its members if no wrongful conduct attributable to the community ([14]). In the same perspective, Emer de Vattel asserted that if a king failed to suppress the injuries inflicted on a foreigner by his citizens, he would be responsible as if he had committed the act personally ([9]). Nevertheless, neither Grotius nor Vattel addressed the question of whether states had an obligation to prevent losses sustained during the internal armed conflict.

The 19th-century authors became special as they tried to answer the question of whether a state had a duty to indemnify losses suffered in the internal armed conflict. One such thinker was Pascale Fiore, who argued that when an internal armed conflict broke out, the state had to deploy all available resources to stop the war; that its liability could be engaged if it neglected to carry out this obligation (Fioré, 1868). Brusa also contributed to this discussion, asserting that a state possesses the authority to command and inflict the punishment to the criminal acts committed on its territory; that if the state failed to exercise this power successfully, it had to compensate the victims. Brusa further linked the state duty to protect foreign nationals on its territory with the economic benefits deriving from the arrival of foreigners on its territory, arguing that if the state failed to curb injuries against them its liability could be engaged ([5]). De Bar connected state liability with its duty to ensure the security to foreigners, stating that if aliens suffer any insecurity, the host state has a legal obligation to indemnify them. Nevertheless, in contrast with the previous writer, De Bar maintained that a state has no duty to indemnify any damage caused by NSAGs ([8]).

Additionally, Wiesse also dismissed the doctrine of the state obligation to prevent losses caused by revolutionists. Although Wiesse insisted that the state responsibility could be engaged in case of any fault, including negligence or lack of due diligence to prevent the harm caused by private persons, he pointed out that this responsibility did not apply to the harm caused by rebels (Wiesse, 1898).

Conclusively, in the nineteenth century, there was a clear consensus among writers regarding the idea that states should be held responsible for failing to prevent illegal acts done by private persons. However, the authors had different opinions about the state’s obligation to prevent losses suffered during the internal armed conflict, including those inflicted by rebels. Certain authors argued that states should be accountable for such losses, while others dismissed this argument.

From 1800 to 1900, various states concluded treaties relating to the facilitation of trade and navigation providing for the state liability to prevent injuries caused by NSAGs. But these treaties conditioned the liability on negligence attributable to the host state. Among these treaties, the Treaty of commerce and navigation between Mexico and The Netherlands, dated September 22, 1897 and the treaty between Mexico and France concluded in 1886 may be cited.

On the other hand, other similar agreements completely overlooked this issue, failing to mention a state’s potential liability for not preventing wrongful conducts of NSAGs. These agreements seemed to reject the idea that states had a legal obligation to prevent such conducts. Instead, the agreements emphasized the principle of equality between aliens and nationals and the total non-responsibility of the host state for the damage caused by rebels (Treaty between Peru and Argentina, 1874)¹. Therefore, all in all, analyzing the all relevant treaties concluded in such a period, it may be concluded that from 1800 to 1900, there was no unanimity among treaties in relation to a legal obligation of states to prevent injuries caused by rebels. Some of them endorsed state liability, while others did not.

With regard to state practice, in various disputes between states that occurred between 1800 and 1900 about the subject matter, the issue of the state responsibility for failure to prevent the damage caused by NSAGs insurgents emerged. However, in all of these disputes no state ever admitted having a legal obligation to prevent losses caused by rebels.

For instance, in 1861, an American national residing in China incurred losses when his property was destroyed by NSAGs opposing the Chinese government. He sought the intervention from the United States to request compensation from China. The United States dismissed his claim asserting that no state could be held liable for the damage caused by NSAGs if no fault could be established against it. The United States explained to the victim that this exclusion of liability on the subject matter was a principle of international law ([25]).

Moreover in 1862 during the American civil war, the cargo belonging to the government of Peru was attacked and destroyed by the Confederate NSAGs. Afterwards, Peru sought compensation from the United States. However, the United States dismissed the claim, stating that the US government had done what it could to prevent the damage. In a third dispute, in 1885 a British citizen sustained injuries in Colombia due to the wrongful conduct of rebels who burnt his neighborhood shortly having captured the area. This citizen asked the United Kingdom government to seek compensation from the Colombian government on his behalf. However, the UK government refused to endorse his claim arguing that the Colombian government’s liability could not be established as it was powerless to prevent the damage. In another dispute that took place in 1896, an American citizen in Rio de Janeiro suffered losses when rebels opposed to the Brazilian government caused the damage to his property. He sought the intervention of the U.S. Department of State in his case to request Brazil to indemnify him. The U.S. Department of State refused to entertain this request, arguing that the Brazilian government had no power to prevent injuries ([25]).

Having examined all disputes noted above, it may be concluded that the nineteenth century’s state practice did not support state liability for failure to prevent losses caused by insurgents. Accordingly, customary international law did not support the doctrine of state liability for failure to prevent losses caused by NSAGs. The opinions of some authors who propounded that such liability existed had no support of any empirical evidence.

In relation to judicial decisions, the state liability for failure to prevent injuries caused by insurgents depended on the status of the concerned state. This liability applied to the states supporting NSAGs but was excluded for states attacked by rebel groups. The basic argument for this exclusion was that states attacked by insurgents had no control over insurgents whereas states supporting rebels had such control.

In the Hanna, the US-British mixed claims commission refused to hold the American government liable for failure to prevent injuries caused by the Confederate insurgents during the American civil war. During the war, these NSAGs caused the damage against the assets of John Hanna, a British citizen. When the Commission noted above examined the subject matter, it determined that the American government was not responsible for the damage suffered. To motivate its decision, the commission stated that the US government had no capacity to prevent injuries inflicted by insurgents ([15]).

Similarly, the concept of a state’s lack of capacity to prevent losses caused by NSAGs was highlighted in the 1892 Venezuelan Steam Transportation Company case rendered by the commission between Venezuela and United States. In the case, the commission pointed out that a state was not liable for losses caused by private persons, but that this liability could be initiated if the state willfully refused to exercise due diligence to prevent such losses. Nonetheless, the commission noted that the state could never be held liable for injuries inflicted on its territory if those injuries were caused by insurgents. As a ground of non-liability, the commission mentioned the lack of control by the state over rebels ([24]).

Moreover, the lack of fault on the part of the host state, especially the inexistence of the lack of due diligence became a ground of non-liability. This occurred in the Hormiguero case whereby the claims commission set up between Spain and the United States refused to hold responsible Spain for the losses sustained by the American citizen in the Cuban civil war. The reason mentioned by the commission was that Spain had no control over the insurrection; that it had committed no fault ([7]).

As previously noted, various judicial bodies held accountable states supporting insurgents for failing to prevent losses inflicted by those NSAGs. A pertinent example of this can be found in the Alabama settled in 1872. In the case, the British government was held liable for injuries caused by the Confederate insurgents against the United States as the British government failed to exercise due diligence to prevent NSAGs to use its territory to launch the military operations against the victim state (United States of America v. Great Britain, 1871)².

From the discussions above, we can conclude that, from 1800 to 1900, states attacked by NSAGs were not liable for injuries caused by insurgents whereas states supporting NSAGs were liable for such injuries. This difference arises from the fact that states attacked by insurgents had no power to prevent the wrongful acts of revolutionists while states supporting rebels to attack other states had it.

The codifications of rules concerning the liability of states for injuries caused by private persons from 1900 to 1945 were made under the League of Nations auspices and by private legal institutions. With regard to the League of Nations, in the report drafted by Gustavo Guerrero on the subject matter in 1926, it was pointed out a state was not liable for injuries sustained by aliens on its territory due to acts of riot or civil war. However, the report highlighted that its liability could be engaged in case of the state’s failure to suppress the damage or punish wrongdoers. Nevertheless, the report explained that the legal obligation imposed on a state was restricted to the area that was under its control ([27]). This seems to indicate that wrongful acts committed in the territory controlled by insurgents were not attributable to the state.

Conversely, in its Bases of discussion prepared in 1929 for the Hague conference for the codification of international law, the Preparatory Committee affirmed that a state was liable for the conduct of rebels³. The ground invoked by the Committee was the state’s failure to exercise due diligence to prevent the wrongful conduct. Still, the Committee’s arguments maintained their low legal value since the Bases of discussion in which they were drafted have never been converted into a binding international convention ([3]).

In addition to the League of Nations, the private institutions also drafted rules governing the liability of states for the damage suffered by aliens on their territories. Writing its draft in 1927, the Institute of International Law asserted that while a state is not liable for the damage suffered by aliens in case of political violence, mob or civil war, its liability could be engaged if the fault could be found on its part either due to lack of due diligence to prevent the harm or due to the state’s failure to accord equal protection to aliens as that accorded to its citizens ([10]). Nonetheless, the Institute did not clarify whether the state could be liable for failure to prevent losses caused by NSAGs.

In relation to the private legal institutions, the Havard Law School was among institutions that played a major role in designing rules relating to state liability. The special element with this School is that it conditioned state liability for injuries caused by rebels to the state’s failure to accord justice to the victims of wrongful acts committed by rebels. In 1929 the school asserted that the responsibility of the state could be engaged if the state committed a certain fault, including the failure to accord aliens efficient redress for the harm sustained ([10])⁴.

Indeed, from the early decades of the twentieth century to 1945, various conventions providing for the state’s obligation to prevent injuries caused by insurgents were concluded by the American states. Under these conventions, a state could be held responsible for the damage caused by revolutionary movements if the lack of due diligence could be attributed to it. One such convention was the Convention Relative to the Rights of Aliens, adopted on January 29, 1902, which stipulated that a state was not responsible for the damage suffered by foreigners as a result of the conduct of insurgents during the civil war, unless in case of the failure of the officials of the state to carry out their obligations. However, this convention did not clarify the specific obligations state officials have towards foreigners.

Another pertinent convention was the Convention on rights and duties of states in event of civil strife, signed on February 20, 1928. This convention did not address the concerns of victim states but rather focused on the responsibilities of aggressor states. Under this agreement, each participating state committed to ensuring that no vessels equipped, armed, or modified for military purposes would be present within its territory, which could be used to support rebellion against another state.

From 1900 to 1945, the judicial bodies punished various states for supporting rebels to fight against other states. However, the same bodies refused to order victim states to pay compensation for injuries caused by insurgents. In relation to states supporting NSAGs, Naulilaa may be useful. According to this case, Portugal sought compensation from Germany due to its military attack against the Portuguese army in Angola during the first World War, alleging that the German colonial authorities in Namibia incited the indigenous Angolan militia to take arms against the Portuguese colonial authorities. Although the arbitrators rejected the allegations of Portugal about the alleged incitement from Germany to insurgents, it nonetheless held liable Germany for injuries suffered resulting from its military actions, as it should have considered that such actions would likely incite the local insurrection ([26]).

In terms of victim states, various judicial bodies took contradictory decisions. In Fulda, the mixed claims commission established between Venezuela and Germany in 1903 ordered Venezuela to compensate the damage suffered by the victim in spite of the lack of evidence of negligence on the part of the Venezuelan authorities. In its memorials before the commission, the Venezuelan government invoked force majeure asserting that the military blockade exercised against it by the European powers had made it unable to suppress the rebellion and prevent injuries against Luis Fulda, a German citizen. The Commission dismissed the defense of Venezuela and required it to compensate the damage suffered, citing the clause of the convention establishing the Commission whereby Venezuela consented to compensate injuries incurred in its civil war ([12]). However, the Commissioners established no single fault on the part of the Venezuelan government.

Conversely, in the Aroa Mines case, the claims commission put in place to settle the disputes between the United Kingdom and Venezuela refused to order Venezuela to indemnify losses caused by NSAGs to the British investor during the war in 1903. As a ground of non-liability, the commission noted that there was no fault on the part of the Venezuelan government ([2]).

Based on the cases appraised above, it can be concluded that from 1900 to 1945, states supporting insurgents to fight other states had a legal obligation to prevent the wrongful conduct of NSAGs. In relation to the states against which revolutionary movements were fighting, this study found no enough judicial decisions supporting their eventual liability for failure to prevent injuries caused by such movements.

What is discussed above about state liability for the conduct of NSAGs, since the nineteenth century to the Second World War, concerned traditional international law on responsibility of States for damage done in their territory to the person or property of foreigners. Before 1945 there were two contradicting rules, namely the international standards of justice and the principle of equality between aliens and nationals ([13]).

The first rule was invoked by developed states to justify the diplomatic protection of aliens abroad while the second rule was used as a defense of developing nations to resist against such diplomatic protection. The international standards of justice were considered as safeguards against the violation of international obligations a state had towards foreigners. Accordingly, aliens could enjoy better treatment than nationals of the host state if this state treated badly its own citizens. This position was disputed by weaker nations invoking the principle of equality between aliens and nationals, under the Calvo doctrine. These states considered the diplomatic protection as a violation by powerful states of the principles of non-intervention and of equality between aliens and nationals ([13]).

With the emergence of international human rights law shortly after the Second World War, those rules became obsolete. Indeed, the international human rights law granted to all persons the protection of their rights without any distinction between aliens and nationals. This protection was considered enough whereby foreigners did not need any other special protection compared to the citizens of the host states.

Indeed, from the Second World War to the present day, new bases of international responsibility for supporting states appeared in various courts and literature. These bases concerned the attribution of NSAG conduct, aid or assistance, occupation/control of territory, and breach of a due-diligence duty to prevent. This is discussed in the next paragraphs in detail.

4.2.1. Attribution of NSAG Conduct

Under international law the conduct of NSAGs is attributable to the state if the group acts under the instructions of or under the direction or control of the state ([19]). This was affirmed by the International Court of Justice (ICJ) in the 1986 Military and paramilitary Activities in and against Nicaraguacase (Nicaragua case). In this case, the ICJ determined that the United States provided substantial support to the contras insurgents to fight against Nicaragua. However, the court refused to hold the United States liable for human rights and humanitarian law violations committed by insurgents due to the fact that the United States had no effective control over the military operations in which violations were committed ([23]). However, in the case the court did not provide the sufficient explanation of the concept “effective control”.

The ICJ explained the scope of effective control test in the 2007 Genocide convention case.In this case, Serbia and Montenegro was charged of supporting the Republika Srpska and other insurgents to commit genocide against the people of Bosnia and Herzegovina. In its decision, the ICJ refused to hold Serbia and Montenegro liable for this genocide due to lack of effective control over insurgents. To assert liability, the court contended that the defendant state had to exercise control over each operation in which violations were committed (Bosnia and Herzegovina v. Serbia and Montenegro, 2007)⁵.

Some courts questioned the theory of effective control. In Tadić case, The International Criminal Tribunal for the former Yugoslavia (the ICTY) developed the overall control standard. In this case, the tribunal determined whether acts of the NSAG could be attributed to the supporting state to elevate an internal armed conflict to the international armed conflict. The ICTY stated that the conduct of the insurrectional movement could be attributed to the supporting state if the state had overall control over the movement, whereby the state would have control in relation to the coordination or general planning of military activities of the movement (Prosecutor v. Duško Tadić, 1999)⁶. Nonetheless, both the overall control test and effective control test provide little protection to the victims of acts committed by NSAGs as it is difficult for the victims of these acts to prove that the concerned state had effective or overall control over the group that committed the alleged violations. Strict liability is better to ensure the enhanced protection of the victims of violations committed by NSAGs.

4.2.2. Aid, Assistance or Complicity

Article 16 of the International Law Commission (ILC) Articles on state responsibility deals with the situation of aid or assistance by the state in the commission of an internationally wrongful act. This article provides that a state which “aids or assists another state in the commission of an internationally wrongful act by the latter is internationally responsible for doing so if that state does so with knowledge of the circumstances of the internationally wrongful act”. State liability is conditioned of the knowledge by the assisting state of the violation of international law committed by another state. The rationale behind this condition is that the assisting state is presumed not to be aware of the wrongfulness of acts of another state ([28]). The main challenge is that the ILC Articles have zero article providing for state liability for aiding or assisting NSAGs. This gives little protection to victims of acts committed by NSAGs with the assistance of a foreign state.

Various courts held states liable for assisting NSAGs to commit human rights violations. For instance, this occurred in the 2005 Mapiripánmassacrecase rendered by the Inter-American Court of Human Rights. In the case, Colombia was charged of supporting a paramilitary group, the AutodefansaUnidas of Colombiato commit massacre against the population of Mapiripán, in the Colombian central part. The Court ordered Colombia to compensate the victims due to the assistance provided by the Colombia military to the armed group (Mapiripánmassacre v. Colombia, 2005)⁷.

Apart from aid or assistance, in some cases complicity with the NSAG has been cited as a basis of liability of the supporting state. This occurred in the Ilaşcucase.In this case, the subject matter was not about an innocent country whose territory had been used to attack another state. Instead, the case was about the complicity of the state whereby Russia was held liable for colluding with the separatist insurgents, refusing to prevent them to acquire weapons used in their violations of international law (Ilaşcuand Others v. Moldova and Russia, 2004)⁸.

4.2.3. Occupation or Control of Territory

Article 1 common to the Four Geneva Conventions requires states parties to respect and ensure respect for the conventions. This article has been interpreted as including the obligation states parties to ensure private parties including NSAGs to respect rules of international humanitarian law.

In the 2005 ArmedActivities on the Territory of the Congo⁹ case, the court held Uganda responsible for failure to take all necessary measures to prevent violations of the rebel movement in Ituri region of the Democratic Republic of Congo. The main factor of liability was the status of the occupying power held by Uganda over the area in which violations were committed. For the court, since Uganda had such a status, it had a duty to curb unlawful conduct of its armed forces and those of private entities operating in the area under its jurisdiction ([1]).

Another relevant case is Chiragov rendered on 16 June 2015 by the European Court of Human Rights. In this case, Mr Elkhan Chiragov and other applicants, Azerbaijani nationals from the district of Lachin, alleged that on 17 May 1992 they were forced to flee from their district due to the civil war that opposed the separatists of the Nagorno-Karabakh region and the government of Azerbaijan. They alleged that those separatists were supported militarily, politically and financially by the government of Armenia; that they were unable to return to their homes due to the Armenian occupation. The Armenian government denied those allegations. Still, the Court held Armenia responsible for the damage suffered by the applicants. As a ground of liability, the Court mentioned the control exercised by Armenia over the territory of Nagorno-Karabakh (Chiragovand Others v. Armenia, 2015)¹⁰.

4.2.4. Breach of a Due Diligence Duty to Prevent

From the Second World War to the present day, a certain number of scholars have investigated the obligation of states to exercise due diligence to prevent wrongful actions of private individuals ([22]). In this context, due diligence has been assessed in accordance with the state ability to do its bests to prevent violations ([20]). The scholars noted that states were required to prevent wrongful actions carried out by private individuals, that if they neglect this duty, they could be held responsible. However, these scholars highlighted that this obligation is relative; that it is dependent on a state’s exercise of due diligence to prevent such actions ([21]).

Various courts have held states liable for breach of a due diligence to prevent unlawful acts of private actors. In the 1980 United States Diplomatic and Consular Staff in Tehran, the ICJ asserted the responsibility of Iran for failure to prevent losses caused by armed militants against the US diplomatic and consular personnel ([30]). As a ground of this responsibility the court mentioned the failure of Iran to take appropriate steps to prevent the losses. In this case, the militants in question appeared on the territory that was under control of the defendant state.

Huma rights courts also dealt with the subject matter. In Osman, the European Court of Human Rights held that the UK’s duty to protect the lives of its people included not only the criminalization of the wrongful conduct, but also taking efficient measures to protect people of the real danger faced by the victims due to wrongful acts of private persons (Osman v. The United Kingdom, 1998)¹¹. The court took a similar decision in Kaya, whereby Türkiye was ordered to compensate losses caused by private persons for its failure to take reasonable measures to ensure the safety of the victims (Mahmut Kaya v.Türkiye, Judgment, 2000)¹². Additionally, the same approach could be found in Tradesmencase, in which the Inter-American Court of Human Rights ordered the Colombia government to indemnify injuries caused by the paramilitaries due to the failure of the Colombian armed forces to curb the violence committed against the victims (the 19[17]). Nonetheless, both the European Court of Human Rights and the Inter-American Court of Human Rights have always maintained that the state’s duty to prevent human rights violations of private actors applies only to the area that is under the jurisdiction or control of the state. This seems to indicate that a state has no legal obligation to protect people that are in the area that is under control of NSAGs.

In any case, in relation to the liability of the supporting state for failure to exercise due diligence to prevent wrongful acts committed by NSAGs, this kind of liability provides insufficient protection to the victims of human rights violations committed by NSAGs. Requiring victims to prove that a certain state was negligent to prevent injuries caused by NSAGs seems difficult as the defendant state has control over the key information involving its liability. Accordingly, for humanitarian reasons, this work suggests the adoption of an international convention providing for strict liability of supporting states for violations committed by NSAGs.

4.3.1. The Search of a Relevant Convention

In relation to the relevant convention about the liability of the attacked state, since the end of the Second World War and the establishment of the United Nations (UN), this latter continued the task that had already been started under the League of Nations in relation to the rules governing state responsibility. In the 1957 Draft Convention on the International Responsibility of the State for Injuries Caused in its Territory to the Person or Property of Aliens, the Special Rapporteur on the subject matter pointed out that negligence by the attacked state to take efficient measures to prevent injuries suffered during the internal armed conflict could lead to state liability (Document A/CN.4/106, 1957). Nonetheless, the Special rapporteur cited no strong legal basis to support his argument.

Instead, in 2001, James Crawford, another special rapporteur, focused on the subject matter. In his commentary to the Draft Articles on State Responsibility, Crawford concurred with the general rule of non-liability, but contended that state liability could be engaged if any fault could be successfully charged against the state, including the failure to prevent or punish wrongful conducts of revolutionists ([18]). Still, Crawford cited no sufficient legal basis supporting his opinion.

Additionally, after the Second World War, private bodies also made drafts discussing state responsibility for losses caused by private persons. In the 1962 Principles of International Law that Govern the Responsibility of the State, the American Juridical Committee indicated that a fault of officials of the country in relation to the damage caused by private individuals could make the country responsible for such damage. However, the Juridical Committee failed to answer the question of whether or not states were responsible for failure to prevent losses caused by NSAGs.

Indeed, the state’s obligation to prevent human rights violations is widely found in various conventions relating to the punishment of certain crimes. These include the 1948 Genocide convention¹³, the 1984¹⁴ torture convention as well conventions relating to the suppression of terrorist activities (Convention on the Safety of United Nations and Associated Personnel, 1994)¹⁵. Although these conventions have a broad language on the duty of state parties to prevent the criminal activities under discussion on their territory, they are quiet about the issue of jurisdiction. In contrast with the international covenant on civil and political rights¹⁶ which restricts the state’s obligation to the territory on which it has jurisdiction or control, the conventions on the prevention and punishment of international crimes noted above state nothing about the subject matter. Importantly, they fail to state anything about any eventual liability of the state for the failure to prevent criminal activities committed by insurrectional movements.

The Geneva conventions of 1949¹⁷ and the Additional Protocol I relating to the protection of victims of international armed conflicts obligate states parties to require their military commanders to prevent breaches of international humanitarian law which would be committed by soldiers under their command. Still, the victims of the non-international armed conflicts seem less protected as the Additional Protocol II relating to such conflicts provides nothing about the duty of military commanders to prevent the same violations. Above all, all of those international humanitarian law instruments are silent about the state’s obligation to prevent the violations done by NSAGs.

International human rights instruments may be cited as legal bases for the attacked state liability to prevent the damage caused by NSAGs. Article 2 of the International Covenant on Civil and Political Rights (ICCPR)¹⁸ requires each state party to ensure to all individuals within its territory and subject to its jurisdiction the rights recognized in the Covenant. The duty “to ensure human rights” was interpreted by courts as a positive obligation of the state party to take positive measures to prevent third parties to violate them (Velásquez-Rodríguez v. Honduras, 1988)¹⁹. Article 2 of the ICCPR noted above may be cited as a legal basis to hold the attacked state liable for injuries caused by NSAGs. However, to be liable, the state needs to have jurisdiction or control over the area in which attacks have been committed. Without this control, the attacked state would successfully invoke force majeure as a defense to dispute its liability.

4.3.2. Conditions of Liability

The question has been whether there is opiniojuris or customary international law supporting the liability of the attacked state for injuries caused by NSAGs. The International Committee on Red Cross (ICRC) gave a positive answer. In its alleged customary international humanitarian law rule 149, the ICRC stated that “a state is responsible for violations of international humanitarian law attributable to it, including those committed by persons or groups acting in fact on its instructions, or under its direction or control”. According to this alleged rule the liability of the state for the conduct of persons acting under its instructions, control or direction, including NSAGs, is part of customary international law. However, the ICRC did not provide enough state practice on the subject matter. Therefore, since there is no enough state practice or opiniojuris on the issue under discussion, it may be asserted that customary international law cannot be cited as a basis of liability of the attacked state for injuries caused by NSAGs.

In its Commentary to Draft articles on Responsibility of States for Internationally Wrongful Acts of 2001, the International Law Commission (ILC) asserted the liability of the attacked state for the conduct of NSAGs if the state committed a certain fault. The fault mentioned by the ILC was the failure of the state to take various measures of vigilance to prevent or punish the wrongful act of the movement ([18]). This was affirmed by the arbitral tribunal in Ampal-American Israel Corpcase whereby Egypt was held liable for failure to prevent the attacks from the armed militants against the oil pipeline of the investor in Northern Sinai. In its decision, the tribunal contended that the standard of liability was not strict liability; that the standard of liability required was due diligence to protect the claimant’s investment. However, it has to be noted that the tribunal held Egypt liable because the militants’ attacks were committed in the area where the state had jurisdiction and control. Although Sinai was characterised by general lawlessness, the state retained some capacity to act and take all necessary measures to prevent the loss (Ampal-American Israel Corp v. Egypt, 2017)²⁰. Accordingly, in this case, the liability of the attacked state was based on two conditions, namely the exercise by the attacked state of control over the area in which wrongful acts were committed and the failure of the state to exercise due diligence to prevent the attacks.