1. Introduction

Reading the Nuhanovic decisions of the Dutch courts, one cannot fail to be struck by the terrible human tragedy that resulted from the decision of Major Franken, deputy commander of Dutchbat, not to include Hasan Nuhanovic’s brother Muhamed (and their mother Nasiha) on the list of the Bosnian civilians allowed to stay in Potocari compound, together with the UN peacekeepers and their local personnel. Having been authorized to stay because he had been part of a civilian committee, one can barely imagine the agonizing choice that Hasan Nuhanovic’s father, Ibro, had to face: stay with his elder son under the protection of Dutchbat for which Hasan worked as an interpreter, or leave the compound and face his destiny, together with his wife Nasiha and younger son Muhamed, in the genocidal hands of the ruthless thugs commanded by Mladic – and then, years later, in the comfort of his study, a scholar is asked to assess whether the Dutch judges correctly applied the customary rules of international law on the attribution of conduct in order to decide who, from the Dutch State, the United Nations, or both, is legally responsible for such tragedy.

While it is a lawyer’s job to reconstruct and make sense of the past, using our painfully learnt grammar, one must admit that the judicial determination of the content of its unwritten customary rules is very often influenced by what can be called, for better or for worse, a certain sense of ‘justice’. If law always simplifies reality and tends to reduce it to a readable, rational, story, good and useful rules are those that allow, as far as possible, for a fine-tuned approach and fair account of the complexity of reality, allowing the settlement of legal disputes in a way which can be said to be consonant with the various moral values and political choices underlying every legal order, of which legal consistency is part.

In order to assess the contribution of the Dutch courts’ decisions to the issue of dual attribution of conduct when national armed forces are lent to the UN for the purpose of peacekeeping operations (3), the various judgments in the Nuhanovic case need to be shortly recalled (2).

2. Three Judgments

a) The District Court

The District Court of The Hague considered that the conduct of members of Dutchbat was to be attributed to the United Nations exclusively. The reason given for such decision is that when a State consents to place its organs ‘under the direction and control’ of another State or, by analogy, of an international organisation, the conduct of those organs is deemed, as a matter of principle, to be attributable to the latter[1]. In affirming this principle, the District Court referred to the ILC’s Articles on the Responsibility of States for Internationally Wrongful Acts (hereinafter ARSIWA), but it actually failed to discuss precisely any of them. Moreover, applying ARSIWA ‘by means of analogy’ to the situation of lending organs to the UN, the judgment seems to somehow confusedly mix the issue of the organ placed at the disposal of another State (Art. 6 ARSIWA) with the issue of the conduct of a person, or a group of persons, directed or controlled by a State (Art. 8 ARSIWA), even if the very notion of being ‘at the disposal’ of a borrowing State requires the latter’s direction and control over the foreign organ[2]. Because Dutchbat was part of UNPROFOR, which had been established by the UN Security Council and was operationally commanded and controlled by the UN[3], the UN was said to be responsible for any of its conduct, even gross negligence or ultra vires actions, unless ‘the State cut[s] across the United Nations command structure’ by ordering a specific wrongful conduct to its troops[4]. In case of ‘parallel’ (i.e. similar) instructions received from the UN and from the lending State, exclusive attribution to the UN would still be the rule[5].

The District Court admitted that the UN and the lending State could specifically agree that the latter be liable to the former in case of gross negligence by members of the deployed troops[6]. The District Court did not elaborate on such hypothesis, but its judgment seems to suggest that in the absence of such contractual provision, the UN would have no right of recovery against the contributing member State, while any such agreement would not alter the attribution rule. In other words, the judgment suggests that the agreement by which a State accepts liability vis-à-vis the UN for any harmful outcome resulting from the gross negligence of its contributing troops is a purely contractual arrangement that does not create any right for third parties, notably the victim of such gross negligence. Even in such a case, the respondent to the victim’s claim would still exclusively be the UN[7]. As stated above, in the opinion of the District Court, the only exception to attribution to the UN is when national authorities ‘cut across the United Nations command structure’ by instructing their troops ‘to ignore UN orders or to go against them’ and those national instructions are carried out. In such a situation, the District Court appears to favour exclusive attribution to the State.

b) The Court of Appeal

The Court of Appeal of The Hague reversed the judgment of the District Court. It found that the Netherlands was responsible for the conduct of the Dutch peacekeepers, because they were under the ‘effective control’ of the State. The Court admitted that Dutchbat was placed under UN command and operated within the UN chain of command, but considered that ‘the decisive criterion for attribution is not who exercised ‘command and control’, but who actually was in possession of ‘effective control’’[8]. Quoting Article 6 of the (then) draft Articles on the Responsibility of International Organizations (hereinafter ARIO) (renumbered Article 7 in ARIO (A/RES/66/100)), the Court of Appeal read the provision in what can be called a ‘reciprocal’ way, as it ruled that:

‘Although strictly speaking this provision only mentions ‘effective control’ in relation to attribution to the ‘hiring’ international organization, it is assumed that the same criterion applies to the question whether the conduct of troops should be attributed to the State who places these troops at the disposal of that other international organization’[9].

Applying the ‘effective control’ criterion in light of the circumstances of the case, the Court of Appeal found that if the Dutch government had instructed Dutchbat ‘not to allow Nuhanovic (as well as his father Ibro Nuhanovic) to leave the compound or to take him along respectively, such an instruction would have been executed’[10]. It concluded therefore that the conduct of Dutchbat was attributable to the State.

Because Hasan Nuhanovic’s claim had been directed solely against the State of the Netherlands, and not also against the United Nations as in other instances related to the genocide committed at Srebrenica[11], the Court of Appeal decided not to enquire about the possibility that the UN had also effective control over the alleged wrongful conduct. The Court nevertheless considered ‘that the possibility that more than one party has ‘effective control’ is generally accepted, which means that it cannot be ruled out that the application of this criterion results in the possibility of attribution to more than one party’[12].

c) The Supreme Court

In conformity with the conclusions presented by the advocate-general[13], the Supreme Court (Hoge Raad) dismissed the cassation appeal. It ruled that the Court of Appeal did not err when it considered that dual attribution through the criterion of effective control was possible, while such criterion did not require that ‘the State [had] countermanded the command structure of the United Nations by giving instructions to Dutchbat or to have exercised operational command independently’[14]. The Supreme Court considered that Dutchbat’s disputed conduct can be attributed to the State, on the basis of Article 7 ARIO and ‘partly in view of what is provided in the attribution rule of Article 8 [ARSIWA]’[15].

3. Appraisal

Assuming, for the sake of this commentary, that Article 7 ARIO reflects international customary law, it appears that the way the higher Dutch courts understood and applied the requirement of ‘effective control’ under that provision is puzzling for at least three reasons.

The first reason is that they assessed effective control of the Dutch authorities over the alleged wrongful conduct, while, for the purpose of Article 7 ARIO, effective control is a matter to be checked on the part of the borrowing organization only (b).

The second reason is that they inferred effective control from hypothetical national orders, i.e., in essence, from a wrongful abstention by the State (c).

The third reason lies in the affirmation that effective control can be multiple and, hence, that dual attribution of conduct would be ‘generally accepted’[16] (d).

Before turning to each of these reasons, the apparent opposition between ‘cutting across’ orders mentioned by the District Court and ‘effective control’ preferred by the Court of Appeal and the Supreme Court deserves some comments (a).

a)  ‘Cutting across’ orders v ‘effective control’

The outcome of the District Court’s decision is perfectly in line with the United Nations Secretariat’s position according to which

‘forces placed at the disposal of the United Nations are “transformed” into a United Nations subsidiary organ and, as such, entail the responsibility of the Organization, just like any other subsidiary organ, regardless of whether control exercised over all aspects of the operation was, in fact, “effective”’. [17]

However, the reasons given by the District Court to affirm the exclusive responsibility of the UN are not based on the structural ‘institutional’ approach of the UN Secretariat which seems to derive a principle of attribution for the purpose of the law of responsibility from Article 29 of the UN Charter conferring to the Security Council the power to establish ‘subsidiary organs’ in order to assist it in the performance of its functions. By using the criterion of ‘direction and control’, the District Court appears to require a more concrete analysis of the facts. However, it actually deduces attribution from the sheer existence of ‘direction and control’ as a matter of principle, therefore coming very close to the UN Secretariat’s institutional approach: the fact of consenting to UN direction and control over national troops has the effect of somehow ‘transforming’ those troops into parts of a UN subsidiary organ, therefore becoming UN organs themselves. Consenting to UN direction and control is not different from putting troops at the disposal of the organisation for the purpose of its subsidiary organ.

It is therefore not so much on the principle, but on the existence of an exception, that the judgment of the District Court appears to differ from the UN Secretariat’s doctrine. As summarized above, in the opinion of the District Court, if the troops obey a national order which ‘cuts across’ the UN control and command structure, the fiction of their ‘transformation’ into UN organs does not hold. In other words, when interfering national orders are obeyed, ‘direction and control’ belongs to the State and the resulting conduct is attributable to it, rather than to the UN.

It is not certain that the ‘cutting across’ exception would be shared by the UN Secretariat, as it holds that the organization remains responsible for any conduct of its peacekeepers, even ‘where the UN command and control structure has broken down’[18]. However, such principled position seems rather artificial in the event of a UN debacle. Since contributing States always retain ‘full command’ over their national troops, thereby allowing them to put them at the disposal of the UN or to terminate their participation in any UN operation, it would obviously be wrong to hold the UN responsible for the conduct of troops that are no longer at its disposal, as a result of a ‘cutting across’ full command order to stop any participation in the UN operation. If that is the case, then one should also admit that national troops remain at the disposal of the UN only in a very formal way when the UN chain of command has effectively collapsed. In such a situation, as the UN no longer is in a position to ‘dispose’ effectively of the lent armed forces, any national order will inevitably ‘cut across’ the vanished UN command structure. The conduct of the obedient soldiers should then be attributed to their State, rather than to the UN. One should also consider that, irrespective of any national orders, the conduct of State organs participating in a UN mission should be attributed to their State when, as a result of the collapse of the UN command and control structure, the UN is no longer in a position to exercise any factual control over them. As will be argued below, that conclusion is warranted because in the absence of effective control by the borrowing organization, attribution to the State of the conduct of its organ is the default customary rule, no matter how much or how little State control is factually exercised.

Despite their differences in criteria and outcomes, there appears to be some similarity between the approaches of the lower and higher Dutch courts. Indeed, by conceiving the ‘cutting across’ national order as a limit to the structural ‘direction and control’ criterion, the District Court seems to share to a certain extent the ‘effective control’ criterion preferred by the Court of Appeal and the Supreme Court. But the District Court only envisages ‘cutting across’ national order as shifting attribution if it is positively interfering with UN command and control, and provided it is actually obeyed. In contrast, in order to establish effective control by the Dutch authorities, the Court of Appeal is satisfied with finding that had they ordered Dutchbat not to do what it did, they would have been obeyed. In other words, while in the opinion of the District Court, ‘effective control’ through ‘cutting across’ orders must not be hypothetical but real in order to defeat general direction and control by the UN resulting from the consent to place troops at its disposal, the Court of Appeal appears to infer ‘effective control’ from the sheer possibility of ‘cutting across’ orders, even if such orders never materialised. The Supreme Court upheld the judgment of the Court of Appeal by affirming that ‘[f]or the purpose of deciding whether the State had effective control it is not necessary for the State to have countermanded the command structure of the United Nations by giving instructions to Dutchbat or to have exercised operational command independently’[19]: by ruling out the requirement of actual ‘cutting across’ orders, such finding does not exclude that ‘effective control’ be inferred from the sheer possibility of such orders.  But if ‘effective control’ of UN peacekeepers by national authorities can be assessed hypothetically, then such control should be assumed as a matter of principle in every situation, since ‘cutting across’ orders are always possible – and certainly today, when satellite and other type of communications allow troops to remain so easily in touch with their national authorities. The added value of effective control understood hypothetically is unclear.

b) Effective control under Article 7 ARIO

As explained in the ILC Commentary, when State organs are ‘fully seconded’ to an organization, they are to be considered not as State organs anymore, but as organs of the latter organization ‘only’[20]: in such cases, Article 6 ARIO commands attribution to the organization, and the situation envisaged under Article 7 ARIO, where State organs are merely ‘placed at the disposal’ of the organization, does not occur.  In other words, and as the very text of Article 7 ARIO makes clear, the situation it governs is predicated on the fact that the person put at the disposal of the organization has not become one of its organ or agent within the meaning of Article 6 ARIO by ‘full secondment’, but is still ‘an organ of a State’ (or of another lending organization[21]):

‘The conduct of an organ of a State or an organ or agent of an international organization that is placed at the disposal of another international organization shall be considered under international law an act of the latter organization if the organization exercises effective control over that conduct.’

Under Article 7 ARIO, the conduct of a State organ, which should in normal circumstances be attributed to that State (under Article 4 ARSIWA), is exceptionally attributed to an international organization because it is ‘placed at [its] disposal’ and because ‘the organization exercises effective control over that conduct’. Disposal and effective control are cumulative conditions for the purpose of the attribution of conduct rule enshrined in Article 7 ARIO: if both are met, the conduct of the State organ is considered as an act of the borrowing organization. Does it mean that such conduct can no longer be considered as an act of the lending State? That issue will be addressed later in this short paper (see d)). What is clear however from the text and logic of Article 7 ARIO is that, in the absence of effective control by the organization over the conduct of the State organ placed at its disposal, such conduct must be considered as an act of that State, and of that State only. There should actually be no need to assess who, from the State or the organization, has exercised effective control over the conduct of the borrowed State organ: what matters is to check if the organization had such control. Ifthe organization did not, the State would be responsible, even if it did not exercise effective control over the conduct of its organ and despite the fact that the organ was nominally at the disposal of the organization. It is indeed superfluous to assess whether the State exercised effective control since the person placed at the disposal of the organization is its organ and that State responsibility for conduct of organs is not conditioned by the positive assessment of any effective control by the State over the conduct of its organ.

Therefore, the sort of ‘reciprocal’ reading of Article 7 ARIO by the Court of Appeal (upheld by the Supreme Court), according to which the criterion of effective control ‘applies to the question whether the conduct of troops should be attributed to the State who places these troops at the disposal of that other international organization’[22], is unwarranted. It is true that the ILC Commentary suggests such reading by emphasizing that

‘The criterion for attribution of conduct either to the contributing State or organization or to the receiving organization is based according to article 7 on the factual control that is exercised over the specific conduct taken by the organ or agent placed at the receiving organization’s disposal.’[23]

However, the ‘factual test of actual control over the conduct in question … is to be applied in the relationship between the organization and the lending State’[24], not in the relationship between the lending State and its own organ placed at the disposal of the organization. It is therefore wrong to think that ‘factual control’ by the State or by the organization is the ‘criterion of attribution… either to the contributing State or… to the receiving organization’: factual control by the organization is the criterion of attribution to the organization of the conduct of a State organ put at its disposal; in the absence of such control by the organization, the default attribution rule applies, so that the conduct of the lent State organ remains attributed to the State, even in the absence of proof of factual control by that State. The reference by the Supreme Court to Article 8 ARSIWA[25] is misplaced.

c) Effective control and wrongful abstention

The assessment by the Court of Appeal of effective control by the Dutch authorities over Dutchbat at the time of the events is based on several elements relating to the decision to evacuate the refugeesand the troops present at Potocari, and the manner in which such decision was carried out. In part, the Court found that the instruction given by Minister Voorhoeve to prevent a separate treatment of the men during the evacuation was not adequately complied with. But more fundamentally, as recalled above, it appears that the Court of Appeal concluded that the Netherlands had retained effective control over the disputed conduct of Dutchbat in the person of Major Franken because ‘it would have had the power to prevent the alleged conduct if it had been aware of this conduct at the time’[26].  Not only does this finding consist, as mentioned above, in inferring effective control from hypothetical events, rather than from events which actually occurred, but it somehow boils down to considering that the Dutch authorities wrongfully abstained from giving orders they should have given, or from issuing instructions countering a specific conduct.

This way of inferring effective control from wrongful abstention for the purpose of attribution is not fully convincing. First, as already argued above, if effective control by national authorities can be assessed hypothetically as a result of a failure to act, then such control should be assumed as a matter of principle in every situation because ‘cutting across’ orders are always possible. Second, the fairness of such finding is doubtful: as the Court of Appeal admitted[27], the Dutch authorities were not aware of the disputed conduct at the time of the events. It is therefore awkward to consider that they should have countered it. Third, and more fundamentally, a finding of wrongful abstention on the part of the Dutch authorities is not conclusive of any attribution issue as far as the conduct of Dutchbat is concerned, as it only concerns the Dutch authorities themselves: ‘[b]y definition, a failure to act never raises any question of attribution, not even ‘negatively’; pointing out a failure to act requires one to identify who had to act, so that the ‘subjective’ element at stake in the search for attribution is always satisfied by finding the wrongful omission.’[28] In other words, pointing at a failure to act is pointing at a possible wrongful conduct of the legal subject bound by an obligation to act, eventually in the form of a duty to prevent certain events that occurred. It is not conclusive of the attribution to that subject of the actual disputed conduct which should not have occurred.

This difficulty brings to light a crucial element that could have been more clearly dealt with by the Dutch courts: what is the precise disputed wrongful act susceptible of triggering the responsibility of the Netherlands? If it is the failure to instruct Dutchbat not to proceed with the evacuation of the Potocari refugees, then no issue of attribution of conduct of the peacekeepers to the UN or to the State is at stake, as the disputed conduct is an omission by the Dutch authorities themselves. It is only if the impugned act is an actual conduct of a member of Dutchbat that the issue of attribution within the meaning of Article 7 ARIO arises. Reading the judgments, it is difficult to assess which internationally wrongful act was specifically contested. The overall impression is that a tragic harmful outcome resulted from a ‘situation’ in which multiple actors were involved. The truth of the matter is, indeed, that not a single conduct or omission caused the injury. But if the impugned act is the decision of Major Franken relating to the separation of the Nuhanovic family, deciding the case on the basis of Article 7 ARIO could have been more straightforward: as he was not ‘fully seconded’ to the UN but only placed at its disposal, his decision must be legally deemed to remain a decision of the State of which he was an organ, unless it can be convincingly established that the UN (through its own agents and organs, or possibly through member States’ organs ‘fully seconded’ to it) had effective control over that specific conduct. Because such control was doubtful in the confused circumstances of the case, finding the responsibility of the Netherlands for Major Franken’s decision is warranted. Judging whether, because of that conduct, the Dutch State should be held responsible for the entire injury suffered by the claimant is another matter that this short paper can only mention, but not discuss.

d) Multiple effective control and dual attribution

As already recalled, the Court of Appeal declared to

‘adopt[…] as a starting point that the possibility that more than one party has ‘effective’ control is generally accepted, which means that it cannot be ruled out that the application of this criterion results in the possibility of attribution to more than one party’[29].

Although the Supreme Court concurred in a more detailed way[30], both courts declined to rule on the eventual effective control by the UN, as the organization was not respondent in this case. In other words, it appears that the affirmation according to which attribution through the test of effective control can be dual is of the nature of an obiter dictum and did not command the judicial outcome of this case.

As stated in the ILC Commentary to which the Supreme Court referred, it is certainly correct to consider that ‘[a]lthough it may not frequently occur in practice, dual or even multiple attribution of conduct cannot be excluded.’[31] As noted by the ILC, this may occur as a result of the conduct of a joint organ. It may also happen when a conduct attributable to a State is nevertheless ‘acknowledged and adopted by an international organization as its own’ within the meaning of Article 9 ARIO – which can be another way to explain the above-mentioned position of the UN Secretariat when it comes to the conduct of peacekeepers. This being said, it is far less certain that attribution of conduct under Article 7 ARIO can be multiple and not exclusive. Because both higher Dutch courts were somehow misled in their analysis of the criterion of effective control over the conduct of lent organs by considering that it should also be applied to the lending State and not only to the borrowing organization, it is understandable that they considered that the application of Article 7 ARIO may result in dual attribution. However, as made clear above, that reading of Article 7 ARIO is clearly not commanded by the ordinary meaning of its terms and its logic, so that effective control must only be checked on the part of the borrowing organization in order to ascertain that the conduct of a State organ placed at its disposal should rather be considered as the organization’s conduct. Moreover, neither the introductory ILC Commentary quoted by the Supreme Court, nor the detailed Commentary of Article 7 ARIO suggests that the application of this provision may result in dual attribution. On the contrary, the ILC Commentary clearly refers to an alternative outcome:

‘The criterion for attribution of conduct either to the contributing State or organization or to the receiving organization is based according to article 7 on the factual control that is exercised over the specific conduct taken by the organ or agent placed at the receiving organization’s disposal’.[32]

In other words, the criterion of effective control by the borrowing organization is envisaged under Article 7 ARIO as a factual element ‘shifting’ attribution from the lending State to the said organization. Even if the wording of Article 7 ARIO does not expressly refer to exclusive attribution of conduct, its drafting would be very poor indeed if its meaning were that its application could result in dual attribution. Therefore, the reference made in the Supreme Court’s judgment to Article 48(1) ARIO[33] appears misplaced if it signifies that the impugned decision of Major Franken is ‘the same internationally wrongful act’ for which both the UN and the Netherlands are responsible as a matter of attribution of conduct on the basis of Article 7 ARIO.

Again, this does not mean that multiple attribution of conduct can never exist. What is argued here is that Article 7 ARIO is a provision which is designed to help identify one responsible entity, not several, the very notion of effective control being exclusive rather than cumulative. Moreover, dual or multiple attribution of conduct is the exception[34] rather than the default position in international law[35]. Of course, attribution of conduct is not exclusive of attribution of responsibility, while a single harmful outcome can be the result of several wrongful acts for which several entities may bear responsibility, either as a matter of attribution of conduct or as a matter of attribution of responsibility. Enquiring on those interplays in the context of the Nuhanovic case, or of UN peacekeeping more generally, is however well beyond the limited scope of this short paper.

[1] Hasan Nuhanovic v Netherlands (Ministry of Defence and Ministry of Foreign Affairs), First Instance Judgment, Decision No LJN: BF0181, Case No 265615, ILDC 1092 (NL 2008), 10th September 2008, District Court, para. 4.8; also available at <http://zoeken.rechtspraak.nl>.

[2] (2001) II/2 YBILC 44, para 2.

[3]Nuhanovic, District Court (n 1), para 4.9.

[4]ibid, para 4.14.1.

[5]ibid, also para 4.14.5.

[6] ibid, para 4.13.

[7] As noted by the ILC in its ARIO commentary: ‘The agreement appears to deal only with distribution of responsibility and not with attribution of conduct. At any event, this type of agreement is not conclusive because it governs only the relations between the contributing State or organization and the receiving organization and could thus not have the effect of depriving a third party of any right that that party may have towards the State or organization which is responsible under the general rules.’: ILC, ‘Report of the International Law Commission on the Work of its 63rd Session (26 April-3 June and 4 July-12 August 2011) UN doc A/66/10, 85, para 3.

[8]Hasan Nuhanovic v Netherlands, Appeal judgment, LJN:BR5388, ILDC 1742 (NL 2011), 5th July 2011, Court of Appeal, para 5.7.

[9]  ibid, para 5.8.

[10]ibid, para 5.18.

[11]Mothers of Srebrenica v Netherlands and the United Nations (Neth App Ct), 30 March 2010, [2010] 49 ILM 1021; Mothers of Srebrenica Association et Al v The Netherlands (Sup Ct of the Neth), 13 April 13, [2012] 51 ILM 1327. On the immunity issue of the UN, see Stichting Mothers of Srebrenica and others v The Netherlands, App no 65542/12, (ECHR, Decision on admissibility, 11 June 2013).

[12]Hasan Nuhanovic v Netherlands, Appeal judgment, LJN:BR5388, ILDC 1742 (NL 2011), 5th July 2011, Court of Appeal, para 5.9.

[13] Parket bij de Hoge Raad, Conclusie 3 May 2013, ECLI:NL:PHR:2013:BZ9225, available also at <http://zoeken.rechtspraak.nl>.

[14]Netherlands (Ministry of Defence and Ministry of Foreign Affairs) v Nuhanovic, Final appeal judgment, ECLI/NL/HR/2013/BZ9225, ILDC 2061 (NL 2013), 12/03324, 6th September 2013, Supreme Court [HR], para 3.11.3.

[15] ibid, para 3.13.

[16]Nuhanovic,Appeal judgment (n 12), para. 5.9.

[17] Comments and observations received from the United Nations, UN doc A/CN.4/637/Add.1, 13, para. 3.

[18]ibid.

[19]Netherlands v Nuhanovic, Supreme Court  (n 14), para 3.13.

[20]Report of the International Law Commission UN doc A/66/10 (n 7) 85, para 1.

[21] For the sake of brevity, this commentary will refer to lent State organs, but the reasoning is mutatis mutandis applicable to organs of a third organization.

[22]Nuhanovic v Netherlands, Court of Appeal (n 8), para 5.18.

[23]Report of the International Law Commission UN doc A/66/10 (n 7) 85, para 4.

[24] Comments and observations received from the United Nations, UN doc A/CN.4/637/Add.1, 13, para 1.

[25](n 15).

[26]Nuhanovic v Netherlands,Court of Appeal (n 12), para 5.18.

[27]ibid.

[28] P d’Argent, ‘Reparation, Cessation, Assurances and Guarantees of Non-Repetition’, fn 75, SHARES Research Paper 34 (2014) available at <www.sharesproject.nl>.

[29]Nuhanovic v Netherlands, Court of Appeal (n 12), para 5.9.

[30]Netherlands v Nuhanovic, Supreme Court (n 14), para 3.9.4.

[31]Report of the International Law Commission UN doc A/66/10 (n 7) 81, para 4.

[32] ibid85, para 4 (I underscore). The French version of the ILC Commentary is even clearer: ‘Le critère d’attribution du comportement soit à l’État ou l’organisation qui fournit des ressources, soit à l’organisation d’accueil repose, selon l’article 7, sur le contrôle qui est exercé dans les faits sur le comportement particulier adopté par l’organe ou l’agent mis à la disposition de l’organisation d’accueil’.

[33]Netherlands v Nuhanovic, Supreme Court (n 14), paras 3.9.2 and 3.9.4.

[34] See A Nollkaemper, ‘Dual Attribution: Liability of the Netherlands for Conduct of Dutchbat in Srebrenica’, (2011) 9 J Intl Criminal Justice 1143-1175; A Nollkaemper & D Jacobs, ‘Shared Responsibility in International Law: A Conceptual Framework’, SHARES Research Paper 03 (2011) available at <www.sharesproject.nl>.

[35] See in favour of multiple attribution of conduct F Messineo, ‘Multiple Attribution of Conduct’, SHARES Research Paper 11 (2012) and also ‘Attribution of Conduct’, SHARES Research Paper 32 (2014) available at <www.sharesproject.nl>.