Zoekresultaat - inzien document
- Parket bij de Hoge Raad
- Datum conclusie
- Datum publicatie
- Formele relaties
Arrest Hoge Raad: ECLI:NL:HR:2019:1284
- Bijzondere kenmerken
Mothers of Srebrenica. Overheidsaansprakelijkheid. Internationaal publiekrecht. Gevolgen van ter beschikking stelling Nederlandse militairen aan VN voor aansprakelijkheid Staat; art. 8 DARS; HR 6 september 2013, ECLI:NL:HR:2013:BZ9225 en BZ9228. Overdracht van ‘command and control’ over Dutchbat aan VN; ‘effective control’; handelen 'ultra vires'. Genocideverdrag; rechtstreekse werking? Onrechtmatige overheidsdaad; maatstaf; EHRM 20 december 2011, nr. 18299/03 (Finogenov e.a./Rusland). Recht op leven en verbod op onmenselijke behandeling (art. 2 en 3 EVRM); zorgvuldigheidsnorm van art. 6:162 BW. Wetenschap Dutchbat van ernstig risico voor mannelijke vluchtelingen; redelijkerwijs te vergen maatregelen om dat risico te vermijden. Heeft Staat onrechtmatig gehandeld door bij evacuatie op 13 juli 1995 de afscheiding door Bosnische Serven van de mannelijke vluchtelingen te vergemakkelijken? Was het onrechtmatig om de 350 mannelijke vluchtelingen die op de compound verbleven niet de keuze te bieden om voorlopig daar te blijven? Was er een reële kans dat deze mannelijke vluchtelingen, als hen die keuze was geboden, uit handen van de Bosnische Serven waren gebleven? Hoge Raad doet zelf de zaak af.
- Verrijkte uitspraak
Case no: 17/04567
Hearing date: 1 February 2019 Opinion in the matter of:
The State of the Netherlands (Ministry of General Affairs, Ministry of Defence and Ministry of Foreign Affairs),
(referred to hereinafter as: “the State”)
1. [respondent 1] ,
2. [respondent 2] ,
3. [respondent 3] ,
4. [respondent 4] ,
5. [respondent 5] ,
6. [respondent 6] ,
7. [respondent 7] ,
8. [respondent 8] ,
9. [respondent 9] ,
10. [respondent 10] ,
all of whom reside in Bosnia and Herzegovina
(also referred to collectively hereinafter as: “ [respondents] ”),
11. the Dutch foundation Stichting Mothers of Srebrenica, having its registered office in Amsterdam, the Netherlands,
(also referred to collectively hereinafter as: “the Foundation et al.”).
The Supreme Court is again being asked to render judgment in proceedings that focus on the dramatic events prompted by the fall of the enclave of Srebrenica on 11 July 1995. In 2012, this Court had to render a judgment on the issue of whether the United Nations (UN) could invoke immunity from jurisdiction in the proceedings which the Foundation et al. had initiated against the State and the UN. The Supreme Court held that the UN was entitled to invoke immunity from jurisdiction.1 The proceedings which the Foundation et al. subsequently initiated at the European Court on Human Rights (“ECtHR”), the ECtHR held that the granting of immunity from jurisdiction to the UN did not constitute a violation of Article 6 of the European Convention on Human Rights (“ECHR”).2 The proceedings initiated by the Foundation against the State were then continued, ultimately leading to the judgment rendered by The Hague Court of Appeal on 17 November 2017, which is now being contested in cassation.3 In 2013, this Court rendered two (virtually identical) judgments in the [A] and [B] cases on the issue of whether the State committed a wrongful act on 13 July 1995 by sending [A] and members of [B]’s family away from the compound of the Dutch battalion of the Airmobile Brigade (referred to hereinafter as: “Dutchbat”), where they had been staying at the time. This act was held to be wrongful because on 13 July 1995, Dutchbat already knew about the risks to which those involved would be exposed. These two Supreme Court judgments are of great relevance to the case presently in cassation.4
Briefly summarised, the present case regards the question of whether the State acted wrongfully towards the Foundation et al. in evacuating 300 Muslim men who had sought refuge after the enclave of Srebrenica fell on 11 July 1995 at a mini safe area comprising the Dutchbat compound in Potočari and a nearby area with vehicle halls. In any case, 25,000 refugees, 5,000 of whom were in the vehicle halls, sought refuge in the mini safe area, where conditions were appalling. On 12 July 1995, at the behest of the Bosnian Serbs, buses and lorries arrived to evacuate the refugees. Dutchbat guided the refugees by forming a sort of “sluice” consisting of vehicles, a human chain of members of Dutchbat and a cordon. This evacuation was halted during the evening of 12 July; it resumed the following day and was completed during the evening of 13 July 1995. Afterwards, it became clear that the Bosnian Serbs had separated the men from the other refugees and killed them. Many victims were never found. The Foundation et al. accuse the State of having acted wrongfully by cooperating with this evacuation of the men, knowing that it was sending those men to their deaths.
My opinion in this case is structured as follows. First (at 2), I will extensively discuss the facts of this case as also reflected in the judgment of The Hague Court of Appeal that is being contested in cassation. I have abbreviated this recitation somewhat, however, by not including the full texts of the UN resolutions. I will also provide an overview of the course of the proceedings (at 3). Following this is a discussion of the grounds for cassation, in which respect I discuss the ground for cross-appeal in cassation first because it is the most extensive. The discussion of the ground for cross-appeal in cassation includes the issue of the standard of effective control, to which the aforementioned [A] and [B] judgments are relevant. Another question that arises is whether the Genocide Convention has a direct effect. Afterwards (at 5), I will discuss the principal ground for appeal, which includes a discussion of Dutchbat’s act of forming the “sluice” and the Court of Appeal’s opinion that Dutchbat acted wrongfully by failing to offer the men who were inside the compound on 13 July 1991 to remain at the compound while the women, children and elderly were evacuated. For the sake of legibility, I have inserted subheadings in the discussion of the parts of the grounds for cassation, always indicating which part of the ground is being discussed.
2 The facts
4 Discussion of the ground for cross-appeal in cassation
The reason for discussing the ground for cross-appeal in cassation first is because this ground is the most extensive. The ground for cross-appeal also raises issues that precede the issues raised in the principal ground for appeal.
The ground for cross-appeal consists of eight parts, which are divided into various sub-parts. Parts 1-3 regard the attribution of Dutchbat’s conduct to the State. Parts 4 and 5 regard the effect of the Genocide Convention. Part 6 regards the standard imposed by the Court of Appeal for reviewing the wrongfulness of Dutchbat’s acts. Part 7 regards Dutchbat’s knowledge of the jeopardy in which the Muslim men were being put. Part 8 regards the ruling of the Court of Appeal concerning the causal connection between the acts of Dutchbat and the fate of the men.
After an introduction,is divided into 12 sub-parts directed at paragraphs 12.1, 12.2, 12.4, 12.5, 12.7-12.12, 17.3, 17.4, 26.1, 27.2, 29.2, 29.4, 29.5a and 29.5b of the contested judgment, as well as the findings referred to in part 1.12 (the catch-all ground). The legal complaints in this part relate to the exercise of effective control in an operational sense. In essence, this part argues that effective control may also ensue from a general, all-encompassing instruction by the State that relates to all aspects of the (subsequent) acts of an organ, in which respect all of the facts and circumstances, and the interrelationship between them, must be taken into consideration when assessing that instruction.
ultra vires acts (i.e. acts that contravene fundamental military principles or military criminal and disciplinary codes) must – regardless of whether those acts must or must not be attributed to the UN pursuant to Article 8 DARIO – be attributed to **Dutchbat** pursuant to Article 7 DARIO because of the organic authority that the State had over Dutchbat and the effective control that this authority automatically entailed over the relevant ultra vires acts. In the Foundation et al.’s written explanation, the grounds in part 2, being the most extensive, are discussed first, followed by an explanation of the grounds in part 1.5, after an introduction, can be divided into three sub-parts directed against paragraphs 15.2, 15.3, 16.1, 16.2, 17.2-17.4, 25.1, 26.1 and 27.2 of the contested judgment. In essence, this part argues that Dutchbat’s
General statement on effective control over Dutchbat’s ultra vires acts, or effective control over operational acts as such
I would note the following first and foremost with regard to the discussion of the two parts. In paragraph 3.07 of the aforementioned judgments in the [A] and [B] cases, the Supreme Court held that when determining the unwritten rules of international law which govern the conditions under which conduct can be attributed to a State or an international organisation, a court may look to two regulations drawn up by the International Law Commission of the United Nations (“ILC”): the Draft Articles on Responsibility of States for Internationally wrongful Acts of 2001 (referred to hereinafter as: DARS)6 and the Draft Articles on the Responsibility of International Organizations of 2001 (referred to hereinafter as: DARIO), respectively.7The Supreme Court subsequently held as follows:
“3.8.1 When answering the question of whether the acts of which Dutchbat is accused may be attributed to the State, the most relevant source is the provisions of the DARS, Part One ‘The internationally wrongful act of a State’, Chapter II ‘Attribution of conduct to a State’, Articles 4 and 8 of which, to the extent relevant here, read as follows:
Conduct of organs of a State
The conduct of any State organ shall be considered an act of that State under international law, whether the organ exercises legislative, executive, judicial or any other functions, whatever position it holds in the organization of the State, and whatever its character as an organ of the central Government or of a territorial unit of the State.
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.”
The Supreme Court subsequently held that Article 4(1) and Article 8 DARS indicate that the conduct of Dutchbat may be attributed to the State if Dutchbat must be considered an organ of the State (Article 4(1) DARS) or if, when engaging in such conduct, Dutchbat was actually acting “on the instructions or under the direction or control” of the State (Article 8 DARS).
The provisions of the DARIO are relevant to the question of whether the conduct of Dutchbat may be attributed only to the UN but not to the State, given that the State had put Dutchbat at the UN’s disposal. Articles 6 and 7 DARIO read as follows:
Conduct of organs or agents of an international organization
1. The conduct of an organ or agent of an international organization in the performance of functions of that organ or agent shall be considered an act of that organization under international law, whatever position the organ or agent holds in respect of the organization.
2. The rules of the organization shall apply in the determination of the functions of its organs and agents.
Conduct of organs of a State or organs or agents of an international organization placed at the disposal of another international organization
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.”
Article 7 DARIO indicates that the conduct of Dutchbat may be attributed to the UN if the UN exercises effective control over that conduct. Paragraph 1 of the Commentary to Article 7 DARIO explains the relationship between this provision and Article 6 DARIO, which pertains to the attribution of acts of organs of international organisations themselves. The explanation quoted in paragraph 3.9.3 of this Court’s judgments in [A] and [B] implies that Article 6 DARIO regards the putting of an organ of a State fully at the disposal of an international organisation. Article 7 DARIO regards a situation in which an organ of the State is put at the disposal of an international organisation but still acts as an organ of the State, as is the case with peacekeeping missions in which the State maintains organic authority. It is in such a situation that the problem of attribution of responsibility for the conduct of an organ so disposed arises.
In the judgments in [A] and [B], the Supreme Court considered (paragraph 3.9.4) the remark in the Commentary to Part Two, Chapter II DARIO (paragraph 4), that Articles 6-9 DARIO did not necessarily result in an act being attributed exclusively to an international organisation – which results in the exclusive responsibility of the international organisation – but that these provisions leave the possibility open that an act will be attributed to an international organisation and a state, which in this case would result in dual attribution.
According to paragraph 3.10.2 of the Supreme Court’s judgments in [A] and [B], the attribution rule laid down in Article 7 DARIO (and therefore the effective control standard) applies to the situation in which a State deploys troops within the context of a UN peacekeeping mission, with command and control being transferred to the UN but with organic authority being retained by the sending State. This is the situation in the present case, as this case involves a peacekeeping mission in which both the UN and the State exercised authority over Dutchbat. Given the possibility of dual attribution pursuant to Article 7 DARIO, and thus the possibility that both the State and the UN exercised effective control, the Court of Appeal could have limited itself to the question of whether the State exercised effective control over Dutchbat. After all, even if the UN also exercised effective control, that does not mean that it was exclusively responsible for the conduct of Dutchbat.
In the present proceedings before the Court of Appeal, the debate between the parties included the question which of Dutchbat’s acts the State exercised effective control over in the aforementioned sense. The grounds in parts 1 and 2 relate to the Court of Appeal’s holding in this respect. The issue of primary relevance in assessing those grounds is how the term “effective control” must be interpreted.
In paragraph 3.11.3 of the judgments in [A] and [B], the Supreme Court held that the assumption of effective control does not require the State to have impinged on the United Nations command structure by instructing Dutchbat or to have exercised independent operational command authority. Also according to paragraph 4 of the Commentary to Article 7 DARIO, the essential issue when attributing an act to the sending State or to the international organisation is the factual control over the specific act, in which respect all of the facts and circumstances, as well as the particular context of the case, must be taken into consideration.
Also relevant is the standard laid down in Article 8 DARS, which the Supreme Court referred to in paragraph 3.13 of the judgments in [A] and [B] when assessing the attribution of Dutchbat’s conduct. This is because Article 8 DARS provides that the conduct of a person or group of persons must be attributed to a State if that 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”. According to paragraph 3 of the Commentary to Article 8 DARS, acts that are carried out under the direction and control of a State may only be attributed to that State if that State “directed or controlled the specific operation and the conduct complained of was an integral part of that operation”. As to the question of what degree of control is required for attribution to the State, paragraph 4 of the Commentary refers to the judgment rendered by the International Court of Justice (“ICJ”) in the Nicaragua v. United States of America, case, in which the ICJ developed the effective control test.8
In the Nicaragua case (para. 106), the ICJ concluded that the United States’ support for the contras had taken various forms over the years, including logistic support, information about the location and movement of Sandinista troops, the use of modern means of communication, and the use of broadcasting networks and radar. The ICJ also held that a number of the military and paramilitary operations had been planned and, to that end, the decision was taken to collaborate closely with, or even directly through, consultants from the United States, based on intelligence and logistic support from the United States, particularly by providing a supply plane to the contras. Despite all of this, and despite the fact that, during a certain period, the contras were so dependent on the United States that they could not have continued the most crucial military and paramilitary operations without US support (para. 111), it cannot be assumed that the contras can be equated in all respects with the US (as though they were an organ of the US, paras. 109 and 110). That is not to say, according to the ICJ, that the United States cannot be held responsible for certain, individual violations of international humanitarian law committed by the contras. Nevertheless, when it came to such individual violations of international humanitarian law, the ICJ also declined to attribute responsibility for these to the United States (para. 115):
“The Court has taken the view (…) that United States participation, even if preponderant or decisive, in the financing, organizing, training, supplying and equipping of the contras, the selection of its military or paramilitary targets, and the planning of the whole of its operation, is still insufficient in itself, on the basis of the evidence in the possession of the Court, for the purpose of attributing to the United States the acts committed by the contras in the course of their military or paramilitary operations in Nicaragua. All the forms of United States participation mentioned above and even the general control by the respondent State over a force with a high degree of dependency on it, would not in themselves mean, without further evidence, that the United States directed or enforced the perpetration of the acts contrary to human rights and humanitarian law alleged by the applicant State. Such acts could well be committed by members of the contras without the control of the United States. For this conduct to give rise to legal responsibility of the United States, it would in principle have to be proved that that State had effective control of the military or paramilitary operations in the course of which the alleged violations were committed”.9
In the Bosnia and Herzegovina v. Serbia and Montenegro10 case (referred to hereinafter as: “Bosnian Genocide case”), the ICJ had to decide whether the acts of the Bosnian Serb army (“VRS”) could be attributed to the former Yugoslavia (“VRJ”). To that end, the ICJ assessed (in the same way it did in Nicaragua) whether the VRS could be seen as an organ of the VRJ (Article 4 DARS) on the ground that the VRS was entirely dependent on the VRJ, or whether, pursuant to Article 8 DARS, responsibility could be attributed to the VRJ because the VRJ had effective control over certain acts on the part of the VRS (para. 398). Citing its judgment in the Nicaragua case, the ICJ found as follows with regard to the effective control standard.
“400. (…) in this context it is not necessary to show that the persons who performed the acts alleged to have violated international law were in general in a relationship of “complete dependence” on the respondent State; it has to be proved that they acted in accordance with that State’s instructions or under its “effective control”. It must however be shown that this “effective control” was exercised, or that the State’s instructions were given, in respect of each operation in which the alleged violations occurred, not generally in respect of the overall actions taken by the persons or groups of persons having committed the violations. (emphasis added, A-G)
401. The Applicant has, it is true, contended that the crime of genocide has a particular nature, in that it may be composed of a considerable number of specific acts separate, to a greater or lesser extent, in time and space. According to the Applicant, this particular nature would justify, among other consequences, assessing the “effective control” of the State allegedly responsible, not in relation to each of these specific acts, but in relation to the whole body of operations carried out by the direct perpetrators of the genocide. The Court is however of the view that the particular characteristics of genocide do not justify the Court in departing from the criterion elaborated in the Judgment in the case concerning Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America) (…). The rules for attributing alleged internationally wrongful conduct to a State do not vary with the nature of the wrongful act in question in the absence of a clearly expressed lex specialis. Genocide will be considered as attributable to a State if and to the extent that the physical acts constitutive of genocide that have been committed by organs or persons other than the State’s own agents were carried out, wholly or in part, on the instructions or directions of the State, or under its effective control. This is the state of customary international law, as reflected in the ILC Articles on State Responsibility.
Here, the ICJ expressly distinguishes “instructions” (as meant in Article 8 DARS) from “effective control”, in which respect the latter encompasses the terms “direction and control” as meant in Article 8 DARS. International law literature asserts that the term “effective control” refers to a level of control that is essentially equivalent to that which a State exercises when it issues direct instructions to perform certain acts.11
In paras. 402 and 403 of the judgment in the Bosnian Genocide case, the ICJ rejected the judgment rendered by the Appeals Chamber of the International Criminal Tribunal for the former Yugoslavia in the Tadić case.12 In that case, the Tribunal expressly declined to follow that standard for attributing responsibility which the ICJ laid down in the Nicaragua judgment. The Tribunal found:
“117. (…) The requirement of international law for the attribution to States of acts performed by private individuals is that the State exercises control over the individuals. The degree of control may, however, vary according to the factual circumstances of each case. The Appeals Chamber fails to see why in each and every circumstance international law should require a high threshold for the test of control.
131. In order to attribute the acts of a military or paramilitary group to a State, it must be proved that the State wields overall control over the group, not only by equipping and financing the group, but also by coordinating or helping in the general planning of its military activity. Only then can the State be held internationally accountable for any misconduct of the group. However, it is not necessary that, in addition, the State should also issue, either to the head or to members of the group, instructions for the commission of specific acts contrary to international law.
137. (…) control by a State over subordinate armed forces or militias or paramilitary units may be of an overall character (and must comprise more than the mere provision of financial assistance or military equipment or training). This requirement, however, does not go so far as to include the issuing of specific orders by the State, or its direction of each individual operation. Under international law it is by no means necessary that the controlling authorities should plan all the operations of the units dependent on them, choose their targets, or give specific instructions concerning the conduct of military operations and any alleged violations of international humanitarian law. (…)”.
In its judgment in the Bosnian Genocide case, the ICJ explicitly rejected the overall control standard developed by the Tribunal. Such a standard broadens the scope of a State’s liability far beyond the fundamental principle of international law that a State is only responsible for its own conduct; in other words, for the conduct of persons who, on whatever basis, act on behalf of the State (para. 406). A State is thus responsible for the acts of its de jure or de facto organs and, in accordance with Article 8 DARS, in cases:
“where an organ of the State gave the instructions or provided the direction pursuant to which the perpetrators of the wrongful act acted or where it exercised effective control over the action during which the wrong was committed”.
In the international law literature, the standard of effective control developed by the ICJ regarding the attribution of responsibility for the conduct of military and paramilitary groups has been criticised as being inconsistent with customary international law and with the fundamental principle that States cannot evade their responsibility to other States if, rather than acting via their own organs, they use groups of individuals to commit acts that harm other States.13 Although the ICJ set a very high standard of proof for attribution in the Nicaragua case, the ILC more or less endorses this standard in the Commentary to Article 7 DARS. That standard may be used as a premise in the case presently in cassation, at least when it comes to the attribution of responsibility pursuant to Article 7 DARS.
As an aside, I note in this context that Rule 17(a) of the Tallinn Manual on the International Law Applicable to Cyber Warfare 2.0 provides that cyber operations carried out by non-state actors are attributable to a given State if those operations were carried out pursuant to the instructions of that State or “under its direction or control”. The international group of experts which prepared the Tallinn Manual 2.0 took the view that the term “effective control” as developed by the ICJ in the Nicaragua and Bosnian Genocide cases prescribes the proper scope of the terms “direction” and “control” as meant in Rule 17(a) of the Tallinn Manual 2.0 and in Article 8 DARS.14
In the literature, Dannenbaum contended that the standard of effective control must lie not in the question of who ordered the alleged acts, but in the question of who was in the best position to prevent those acts (the power to prevent standard). 15 Subsequently, Dannenbaum pointed out that the standard of effective control which the ICJ applied in the Nicaragua case is much more stringent.16 Dannenbaum asserts that there are good reasons for believing that the concept of effective control as applied in the Nicaragua and Bosnian Genocide cases deviates from that applied in situations such as peacekeeping missions, which are subject to Article 7 DARIO. This is because the military and paramilitary groups who violated international humanitarian rights in the Nicaragua and Bosnian Genocide cases had no de jure relationship with the US and the VRJ, respectively, while in the context of peacekeeping missions, the troops do always have de jure relationships with the sending State and the UN.17
Although there is indeed a difference between the attribution of the conduct of private persons (including military and paramilitary groups) and the attribution of the conduct of UN peacekeeping forces, the standard of effective control established by the ICJ is relevant to assessing the effective control over the conduct of peacekeeping forces such as Dutchbat.18 The Supreme Court also seems to use this as a premise, given the reference to Article 8 DARS in paragraph 3.13 of its judgments in [B] and [A]. When command and control over peacekeeping forces is transferred to the UN,19 the premise is that the UN, to the exclusion of the sending State, exercises command and control over the operational execution of the peacekeeping forces’ mandate.20 That is also not in dispute in the present case.21 In principle, therefore, the operational actions of UN peacekeeping forces are not actions of the sending State.22 That is why the principle that a State is solely responsible for its own conduct applies here, as well, and why the standard of proof for attributing conduct to a State which does not, either in principle or based on a certain viewpoint, constitute conduct by that State must be high.23 That means that the operational conduct of UN peacekeeping forces can only be attributed to the sending State (naturally in addition to a situation in which the sending State impinges on the UN command structure or independently exercises operational control) if the State, through an active form of control that is directly aimed at a specific operation or operational conduct, obtains factual control over the relevant operation or operational conduct. In this context, it is a given that a general, abstract instruction that does not relate to a specific operation or operational conduct cannot in itself – and thus without any additional, specific control aimed at that operation or conduct – result in effective control.
The foregoing implies that the “power to prevent standard” advocated by Dannenbaum, to which reference is made, inter alia, in paragraph 5.3.7 of the Foundation et al.’s written explanation must be rejected, at least to the extent that standard for attribution to the sending State is broader than that implied by the foregoing.24 The fact that the power to prevent standard must be rejected to that extent also follows from the fact that attribution pursuant to Article 7 DARIO must be effected based on all of the facts and circumstances and the particular context of the case rather than on a broad normative standard for attribution based on an abstraction of the facts of the case.25 In this respect, I also note the following. In the [B] and [A] cases, The Hague Court of Appeal held that when assessing the question of whether a State has exercised effective control it was not only relevant whether the State gave direct instructions to Dutchbat, but also whether the State had it within its power to prevent Dutchbat’s conduct.26 In my view, the latter does not hold true when it comes to attributing responsibility for the operational conduct of peacekeeping forces. The Court of Appeal’s standard was not adopted in paragraph 3.11.3 of the Supreme Court judgments in [B] and [A], but the Supreme Court did find that the issue in question was factual control over the specific conduct, in which respect all of the facts and circumstances as well as the particular context of the case had to be taken into account.
The power to prevent standard may be usable in a different respect, however. As noted, command and control authority in UN peacekeeping missions is transferred to the UN, while the State retains organisational authority as well as certain strategic authority (the authority to withdraw troops and discontinue participation.27 The assumption therefore would be that at least all operational acts would be attributed to the UN. One might argue that the assumption must be that activities that initially fall within the purview of the organisational authority of the State must also be attributed to the State. In that sense, the power to prevent standard may very well be relevant: in principle, the party with legal authority over the conduct would be best positioned to prevent that conduct, and thus it may be assumed that that conduct could be attributed to that party. This does not mean that the situation might be different based on the facts and circumstances and particular context of the case. The power to prevent standard thus functions as a sort of evidentiary rule. If it has been established that a certain act falls within the purview of a certain authority, it may be assumed at the outset that the person exercising that authority has effective control over the act. Incidentally, all of this is consistent with the rationale for dual attribution under Article 7 DARIO. With regard to operational acts by UN peacekeeping troops, however, the sending State generally has no relevant legal authority, which is why the attribution of operational acts to the sending State is subject to the strict standard of effective control set out above. This is not say that it would be impossible for a State’s exercise of its organisational authority to constitute the exercise of effective control over the operational acts of the UN peacekeeping troops. That may very well be the case, but it must always be demonstrated then, based on all of the facts and circumstances and the particular context of the case, that the sending State exercised effective control over the operational act.
What does this mean for ultra vires acts committed by UN peacekeeping troops? Dannenbaum defines ultra vires acts as acts taken “outside of the scope of the action authorized by the U.N. Commander”.28 The mere fact that UN peacekeeping forces so act in the scope of operations or operational acts does not mean that these ultra vires acts must then be attributed to the sending State.29 To the extent such acts remain within “the official capacity and within the overall functions” of the UN (Cf. Article 8 DARIO and paragraph 9 of the Commentary),30 it will continue to be assumed that the UN – to the exclusion of the sending State – is exercising effective control over that act. In principle, therefore, such acts do not constitute acts by the sending State, which means that a high threshold must be met before responsibility for these acts is attributed to the sending State. That need not automatically apply to misdeeds committed by individual troop members when the execution of an operation was under way. What if peacekeeping troops are ordered to patrol and observe in a certain area (operational act) but an individual soldier needlessly and without provocation shoots and kills a civilian during that patrol? One could argue that such acts initially fall within the purview of the organisational authority of the sending State, so that in this case the assumption would be that that act would also be attributed to the sending State.31
After this general explanation, I will return to the discussion of the ground for cross-appeal. As I already noted above,of the Foundation et al.’s written explanation must be considered as the most extensive. For this reason, I discuss the grounds in that part first. The part is divided into three sub-parts.
Part 2 (ground for cross-appeal): ultra vires acts by Dutchbat?
ultra vires acts by Dutchbat indicates a misinterpretation of the law, since the acts of UN peacekeeping troops must always be attributed to the sending State if the acts contravene the UN’s instructions to the peacekeeping troops. contends that the latter applies at any rate to acts that contravene the fundamental military principles that are in any case encompassed by the State’s organisational order. The Court of Appeal thus erroneously failed to examine whether the acts of Dutchbat alleged by the Foundation et al. contravened fundamental military principles. contends that the Court of Appeal failed to appreciate that every act by an organ of a State, and thus also the ultra vires acts of peacekeeping troops, must be attributed to the State unless such acts are committed in the actor’s capacity as a private individual, in which respect this part refers to Article 7 DARS.contends that the Court of Appeal’s holding regarding the attribution of the
From the above, and from what I wrote in paragraph 4.22 in particular, it follows that the Court of Appeal did not base its holding in para. 15.3 – that the ultra vires acts of Dutchbat could not be attributed to the sending State merely because the State had control over troop preparation, personal affairs, and troop discipline or the fact that Dutchbat is an organ of the State – on a misinterpretation of the law. The Court of Appeal correctly assumed that when such operational acts are performed within “the official capacity and within the overall functions” of the UN, they may only be attributed to the State if the State exercised effective control over those acts as meant in Article 7 DARIO. The complaints in parts 2.1 and 2.3 thus fail.
The part also asserts that ultra vires acts committed by individual soldiers may also be assumed to be attributable to the State to the extent it is established that authority retained by the State initially encompassed those acts, thus imbuing the State with the power to prevent those acts. Such a situation does not exist, however, just because those ultra vires acts contravene fundamental military principles. The Court of Appeal saw no need to examine these issues individually. The complaint in also fails.
Part 1 (ground for cross-appeal): effective control of the State?
is divided into 11 sub-parts. contends that in para. 12.1, et seq. the Court of Appeal failed to appreciate that effective control may also ensue from a general, all-encompassing instruction from the State, or may also be shown by the fact that the State was in such a position that it was within its power to prevent the operational acts at issue had it been aware of them in good time. contends that to the extent the Court of Appeal’s holding is based on the premise that a general, all-encompassing instruction from the State cannot be sufficient for effective control if specific instructions from the UN followed later, and the State does not afterwards give any instructions itself, that holding is also legally incorrect.
are premised on a misinterpretation of the law that effective control may ensue from a general, all-encompassing instruction as such. I suffice by referring to the general explanation of effective control which I provide above. The fact that the State could have prevented the operational acts referred to in part 1.1 is not a sufficient basis for assuming that the State had effective control over these acts. The complaints of both parts thus fail.
mutatis mutandis to the statements made by General Couzy on 16 and 23 July 1995, which the Court of Appeal ignored due to the lack of a causal connection (para. 12.11). The Foundation et al. asserted these statements as support for the complaint that the Dutchbat conduct at issue in these proceedings was determined by a general instruction from the State. This part also contends that the Court of Appeal should have taken into consideration the instructions asserted by the Foundation et al. in the context of the argument that the State had control over the (initial) prevention and (later) cessation of air support.contends that the Court of Appeal erred in interpreting the law by examining the Foundation et al.’s various assertions separately instead of in relation to one another. This pertains primarily to the statements referred to in para. 12.5 and the findings in para. 12.8 regarding (i) clearer communication of what the parties want, (ii) informal consultations, (iii) the fact that the UN chain of command included persons with an active role who were also promoting the interests of the Netherlands, and the guidelines and priorities prepared by the Defence Crisis Management Centre (“DCBC”) referenced in para. 12.9. In addition, the Court of Appeal should not have disregarded the Foundation et al.’s offer of proof regarding the situation report because it contained reports of meetings held after the acts at issue in the present proceedings were committed and because it concerned information from Potočari (para. 12.10), since the reports could have made a solid contribution to the control exercised by the State when those acts were committed. This part asserts that the same applies
I note the following regarding the complaints in this part. The Court of Appeal apparently took into account the instructions and circumstances referred to in part 1.3 when assessing the question of whether Voorhoeve’s general, all-encompassing instruction, which in itself did not create effective control, nevertheless had consequences in the form of more specific orders or instructions from which effective control could actually have been derived. The Court of Appeal did assess the aforementioned instructions and circumstances in that context and in relationship to one another. In para. 12.12, the Court of Appeal concluded that no effective control could be derived despite the instructions that casualties had to be avoided and that the troops’ own safety took priority. There is no factual basis for the complaint that the Court of Appeal only considered all of the instructions asserted by the Foundation et al. separately.
In para. 12.10 – which itself is not in dispute in cassation – the Court of Appeal found that the Foundation et al. offered proof for their assertion that General Smith stated that the situation reports would have to be released now and that the confidential situation report contained the following text:
“The initial discussions with the freed Dutch peacekeepers showed that Bosnian men were executed; the number of victims is estimated at 50 to 100. (…) Expectations are that this will not be announced to the world at large until after Dutchbat leaves Potočari to avoid complications”.
I cannot see how these assertions, were they proven to be correct, could contribute to a finding that specific acts were committed pursuant to the instructions of the State. For that reason alone, this complaint in part 1.3 fails.
To the extent this part complains that the Court of Appeal’s assessment relating to the assertions that the Foundation et al. made concerning General Couzy’s statements on 16 and 23 July 1995, the complaint ignores the fact that no effective control can be derived from a general instruction of the State. The Court of Appeal was not obliged, therefore, to examine the extent to which those statements could support the assertion that Dutchbat’s conduct was determined by the general instruction of the State, which means this complaint must also fail.
In paras. 29.1-31, the Court of Appeal assessed the attribution of responsibility for prevention and cessation of air support. When assessing whether effective control over other specific acts ensued from the general, all-encompassing instruction, the Court of Appeal was not bound to consider the instructions asserted by the Foundation et al. regarding the effective control of the specific operational act of preventing and ceasing the provision of air support. After all, effective control over a specific operation or operational act cannot ensue from a general, all-encompassing instruction as such. Furthermore, it must be established whether the required effective control was present for each specific operation or operational act separately. Therefore, the last complaint in part 1.3 also fails.
relates to para. 12.4, in which the Court of Appeal found that Dutchbat’s actual conduct was determined on-site by the specific orders that were issued by the commanders in the UN chain of command and not by the instructions of the State. This part complains that this assessment is incomprehensible, or at least insufficiently substantiated, since the Foundation et al. asserted that many acts by Dutchbat actually contravened the orders issued by the UN chain of command, but complied entirely with Voorhoeve’s general and all-encompassing instruction to place the “very highest priority” on the safety of the Dutch soldiers, and the Court of Appeal provides absolutely no substantiation for which acts were determined by specific orders issued by the UN chain of command.
This part of the ground for cassation fails. In para. 12.6, the Court of Appeal apparently builds on its finding in para. 12.4. In para. 12.6, the Court of Appeal finds that the statements of, inter alia, Nicolai, Voorhoeve, Van den Breemen and Hilderink show that, in practice, everyone was well aware that command and control over Dutchbat was both formally and actually vested in the UN – including on or about 10 July 1995 (in other words, around the time Voorhoeve issued his general instructions). In para. 12.7, the Court of Appeal found that the UN had given Dutchbat specific instructions, including with regard to taking blocking positions on 9 July 1995 (see also para. 2.34). In so doing, the Court of Appeal provided a sufficient explanation for its finding in para. 12.4.
complains that the Court of Appeal based its conclusion in para. 12.4 on a misinterpretation of the law if the Court of Appeal intended to state that the mere fact that the general instruction issued by the UN that priority had to be given to the safety of UN staff was made before the similar instruction by the State means that effective control cannot be derived from the State’s instruction. To the extent that the Court of Appeal’s use of the word “endorse” to express its view that the State was merely “seconding” the UN’s general instruction, there is no sufficient substantiation for that finding, since, according to this complaint, the Court of Appeal provided absolutely no indication of the basis for this finding. contends that para. 12.4 is incomprehensible, or at least insufficiently substantiated, to the extent it holds that the acts that were consistent with the State’s general announcement also fully complied with the general order issued by the UN, such that the performance of these acts does not indicate effective control by the State. After all, the Foundation et al. have asserted that the conduct of Dutchbat at issue in this case contravened the specific instructions that were issued by the UN chain of command after the UN’s general instruction.
can be derived from a general instruction by the State. Moreover, the complaint is based on a misinterpretation of the contested judgment. The Court of Appeal always assessed whether the State had effective control over specific acts committed by Dutchbat and did not condition that assessment on whether the State did or did not “second” a previous general instruction by the UN.fails because it is based on a misinterpretation of the law; it contends that effective control
et seq.).also fails, because it regards an incorrect reading of the contested judgment. After all, the Court of Appeal’s rejection of the Foundation et al.’s assertion that Dutchbat committed certain acts in contravention of the specific instructions issued by the UN chain of command and its complaint that these acts must therefore be attributed to the State was not based on the fact that these acts contravened the UN’s general order, but rather on the Court of Appeal’s own assessment of the specific acts themselves (para. 14,
relates to the Court of Appeal’s finding in para. 12.5 that the statements of Voorhoeve, Van den Breemen and Franken to which the District Court referred cannot support the finding that the State had control of operational affairs or otherwise exercised effective control by means of an instruction. This part refers to many other assertions which the Foundation et al. have invoked (a total of 19 sections are formulated, with sections 18 and 19 being further subdivided into 12 sub-sections). This part complains that the Court of Appeal should not have ignored these assertions by the Foundation et al. without further substantiation.
The Foundation et al. take the position, as it appears from their written explanation (under 12.6.2), that the instructions referred to in the part should have been taken into consideration when answering the question of whether there was effective control in the form of a general, all-encompassing instruction. Since effective control cannot ensue from general, all-encompassing instruction, however, this part fails for that reason alone. In addition, as the State has also noted in paras. 3.2.41 and 3.2.42 of its written explanation, some of the issues raised in this part were already decided in the first instance and cannot be raised again in the Foundation et al.’s grounds for cassation, and, furthermore, not all of the issues raised on appeal were raised in the context of ground for appeal 4 (the assessment of which is the subject of complaint in part 1.6).
For the sake of completeness, I address the issues raised in the sections of this part on a group-by-group basis.
Sub-sections (i)-(vi), (xiii) and (xiv) regard statements that exclusively regard the general instruction to avoid unnecessary casualties and prioritise the safety of the staff. In para. 12.4, the Court of Appeal properly held that no effective control can ensue from that general instruction as such (even to the extent the instruction were to more specifically regard the situation relating to the blocking positions, as asserted in sub-sections (xiii) and (xiv)). After all, the Court of Appeal held in para. 12.4 that general instructions (about avoiding casualties, placing the very highest priority on the safety of the Dutch soldiers, and returning the men and women home in one piece) were not aimed at specific operational acts, such as whether or not to abandon certain observation posts or blocking positions. Therefore, the Court of Appeal was not obliged to give separate consideration to the statements in question. In paras. 12.5 and 12.9, moreover, the Court of Appeal did properly address the statement by Franken referred to in sub-section (i) and the comments made at the DCBC meeting referred to in sub-sections (ii)-(iv),32and the Court of Appeal concluded that effective control did not ensue from any of those statements.
Sub-sections (vii), (x), (xi), (xvi) and (xvii) relate to the statements and facts regarding the transition period (the evening of 11 July 1995), which was the point from which the Court of Appeal held that the State had effective control. The Court of Appeal was thus not bound to address these sub-sections separately. To the extent these statements and circumstances were asserted as an indication that the State exercised effective control by issuing the general and all-encompassing instruction, we must again look to the fact that effective control cannot ensue from that instruction, which means that the Court of Appeal was not required to take separate account of these sub-sections in its assessment.
Sub-sections (viii) and (ix) relate to remarks of a general nature made by Voorhoeve during the Parliamentary regarding the deviation from the division of power between the UN and the State, or at least the State’s having ignored formal agreements on chains of command. Sub-section (xii) regards a passage from the NIOD report concerning the appointment of Nicolai as commander of the Dutch contingent. In para. 12.6 the Court of Appeal held, citing Voorhoeve’s testimony before the Parliamentary Inquiry Commission, that everyone was still very well aware on or about 10 July 1995 that command and control over Dutchbat had been transferred to the UN and issued its findings in para. 23.1, et seq., regarding the extent to which the State exercised control over Dutchbat starting on 11 July 1995. The Court of Appeal therefore acknowledged, contrary to the suggestion in part 1.6, that the formal agreements about chains of command and command and control were deviated from to a lesser or greater extent at some point, and it made a finding as to when that occurred. That finding is sufficiently comprehensible even without referring to the passages of the Parliamentary Inquiry cited in sub-sections (viii) and (ix) and the passage from the NIOD report referred to in sub-section (xii). Moreover, I cannot see how the passages referred to in sub-section (xii) regarding Nicolai’s appointment as commander of the Dutch contingent could contribute to a finding that the State exercised effective control over Dutchbat at any point in time.
Sub-sections (xv), (xviii) and (xix) regard the events surrounding the prevention and cessation of air support and relate to Karremans’ and Nicolai’s attitude concerning that air support, as well as to the influence that Voorhoeve, Hilderink and Van den Breemen had on the decision-making about that air support within the UN and NATO. In paras. 29.1-31 the Court of Appeal made substantiated findings relating to the Foundation et al.’s ground for appeal 12. The Court of Appeal took the aforementioned statements and facts into account in its assessment, in which it held (i) that no facts or circumstances had been asserted based on which it could be established that persons other than those in the UN chain of command and NATO (i.e. persons other than Karremans and Nicolai) exercised effective control over specific requests for air support; (ii) that no assertion had been made that the State interfered with Dutchbat’s request for air support; (iii) that it had not been established that the State obstructed the honouring of the requests for air support; and (iv) that, given Article 59(2) DARIO, the influence which the Netherlands had on the decisions taken by NATO and the UN could not lead to the NATO decision being attributed to the Netherlands.33 In so doing, the Court of Appeal provided a sufficiently comprehensible response to all of the assertions made in sub-sections (xv), (xviii) and (xix), all of which of course regard either the attitude of Karremans and Nicolai or the influence which Voorhoeve, Hilderink and Van den Breemen had on decision-making within the UN and NATO. In this respect, it is worth noting that it is precisely the assertions made in sub-sections (xviii)-c and (xix)-a that support the Court of Appeal’s finding that the State did not obstruct the provision of air support, because those sub-sections quote Voorhoeve’s statements, in which he contended that air support was unavoidable and that close air support had to be provided if the local commander considered such necessary to the safety of the peacekeepers.
The conclusion is that the Court of Appeal was not required to address separately the assertions made raised in this part. The Court of Appeal’s assessment took these assertions into account in a comprehensible manner. Therefore, the complaint contained in part 1.6 must also fail.
complains that the Court of Appeal’s finding in para. 12.6 – that everyone involved in the actual conduct, both in The Hague and at the UN and in Dutchbat, were very well aware that command and control had been transferred to the UN – is incomprehensible since awareness of the transfer of formal control says little to nothing about effective control. Furthermore, according to the complaint, the Court of Appeal’s finding did not expressly take into account either Voorhoeve’s statements which were summarised in subsections (vii), (ix) and (xiii) of part 1.6 or Nicolai’s statements and the passage from the NIOD report cited in subsections (xi) and (xii).
The complaint fails. In the disputed judgment, the Court of Appeal wanted to express that the formal division of powers between the UN and the State was also respected in practice. The Court of Appeal thus did not find that the mere awareness of that formal division of powers already provided an indication of effective control. When discussing part 1.6, above, I explained that the Court of Appeal was not required to expressly address or respond with sufficient substantiation to the statements cited in subsections (vii), (ix), (xiii), (xi) and (xii) of part 1.6.
complains about the Court of Appeal’s finding in para. 12.7 that the fact that Dutchbat committed certain acts after Voorhoeve’s instructions were issued did not mean that those acts were prompted (either in whole or in part) by those instructions. This part complains that the Foundation et al. asserted many specific acts committed by Dutchbat that contravened the orders of the UN and that this indicates that those acts can be attributed to the general, all-encompassing order issued by the State. Given the Foundation et al.’s difficult evidentiary position, this part asserts, the Court of Appeal should have investigated whether the acts at issue in this case contravened UN orders (or other obligations) and should have set forth the results of this investigation in the grounds for its decision.
In paras. 8.2 and 8.3, the Court of Appeal held that there was no reason to permit the Foundation et al.’s duty to plead and prove their case to a generally lower standard when assessing the disputed issues, which meant that there was also no reason to take the Foundation et al.’s evidentiary position into account when assessing the connection between Voorhoeve’s instructions and the later conduct of Dutchbat. Regardless, as I note above in my general explanation and in para. 4.22, ultra vires operational acts cannot be attributed to the sending State merely because those acts are ultra vires, and effective control cannot ensue from a general, all-encompassing instruction. Even leaving aside the fact that the Court of Appeal’s finding in para. 12.7 that a temporal connection does not imply a causal connection was perfectly comprehensible, the fact that the Foundation et al. asserted that many acts committed by Dutchbat contravened UN orders cannot, in itself, lead to the conclusion that the Court of Appeal should have made different findings regarding the connection between Voorhoeve’s instructions and Dutchbat’s later acts. The complaints in part 1.8 thus fail.
relates to the Court of Appeal’s finding in para. 12.8 that there was no effective control based on the fact that persons within the UN chain of command, including Nicolai, also acted as advocates for the interests of the State, and because that fact does not entail that the State was part of the UN chain of command or had such influence on that chain of command that the specific instructions issued by the UN chain of command were actually issued (in part) by the State. This part complains that this finding is either erroneous or incomprehensible because the Foundation et al. asserted that when it came to certain issues, Nicolai was actually acting pursuant to the State’s instructions, in which respect the Foundation et al. noted that, from a certain point forward, Nicolai was also definitely also formally acting as the commander of the Dutch contingent. Therefore, this part asserts, the State exercised effective control over Dutchbat via Nicolai. This part also complains that the Court of Appeal premised its finding on a misinterpretation of the law to the extent that premise is that the State can only be said to have effective control to the extent that the State constitutes part of the UN chain of command or to the extent that specific instructions issued by the UN chain of command must be considered to have actually been issued by the State.
The assertions which this part invokes with regard to the instructions given to Nicolai (with the exception of the general, all-encompassing instruction and the instructions relating to the moment at which the Court of Appeal assumed that effective control existed)34only regard the prevention of air support.35 The Court of Appeal provided a substantiated discussion of the allocation of responsibility for preventing that air support in paras. 29.1-29.4 of its judgment, such that the Court of Appeal did not need to address the assertions in question in para. 12.8 as well. This part also refers (in footnote 59) to a letter dated 12 July 1995 which Karremans sent to Mladić (which letter was also mentioned in part 1.6 (xvi), in which Karremans wrote:
“At 20.00 hrs I did receive a message from the authorities of the Netherlands thru HQ UNPROFOR in Sarajevo concerning the evacuation of Dutchbat. I have been ordered (…)”.
These concern a statement regarding the transition period with effect from which, the Court of Appeal held, the State exercised effective control. The Court of Appeal was thus not bound to address this issue separately. To that extent, the complaint fails.
By ruling that the State did not become part of the UN line of command merely on the basis of the fact that the Dutch were active within the UN line of command, or that it could exert influence to such an extent that instructions given through the UN line of command were actually given by the State, the Court of Appeal did not lose sight of the fact that the State may have also been able to exercise effective control in another manner. A large part of para. 12.1 et seq. actually discusses that very question. The complaint from part 1.9 also fails in this regard.
de facto from his double role on instructions of the State, so that in reality the decisions were not taken within the UN and NATO, and that it was because of Nicolai that the requests for air support did not reach the decision-making levels within the UN and NATO.contains a complaint directed against the Court of Appeal’s ruling in paras. 29.2 and 29.4 regarding the withholding of air support. The part complains that the opinion is incomprehensible because, contrary to the Court of Appeal’s ruling, others than those from the UN line of command and NATO exercised factual control of specific requests for air support, and because facts and circumstances certainly have been put forward that can support the conclusion that the State had effective control over Dutchbat’s requests for air support. In substantiation the part refers to what was asserted in part 1.6 at (xi), (xii) and (xviii), in particular at a, d and e. The Foundation et al. have also argued in this context that Nicolai acted
Nicolai’s statement cited in part 1.6 at (xi), where he says that he had “a bit of a double role”, pertains to an instruction of 13 July 1995, and therefore to the period after the decisions regarding the request for air support had already been taken, and in respect of which the Court of Appeal furthermore assumed effective control. It can only be derived from the arguments in part 1.6 at (xii) and (xviii) that Nicolai took certain decisions regarding air support, also taking into account the interests of the Dutch hostages. However, these points provide no indication that Nicolai was acting based on specific instructions from the State, and not merely in line with the general instruction given by both the State and the UN that the personal safety of Dutchbat had priority.36 The contested opinions in paras. 29.2 and 29.4 are not incomprehensible for that reason, and therefore fail.
In essence,merely repeats the complaints from parts 1.6 and 1.10.1, and therefore shares their fate.
also contain reasoning complaints that reiterate the complaints in parts 1.4-1.10. These complaints therefore also fail.
contains complaints that embroider upon the preceding parts 1.1-1.11 and also share that fate.
The final conclusion is that part 1 entirely fails.
Part 3 (ground for cross-appeal in cassation): decision-making regarding the discontinuation of air support
is directed against paras. 29.6 and 29.8 of the contested judgment, and can be divided into three sub-parts. In para. 29.6 the Court of Appeal held that the influence exerted by the State in NATO’s decisions cannot, in light of Article 59(2) DARIO, lead to responsibility for those decisions being attributed to the State. In para. 29.8 the Court of Appeal held that nothing was put forward that indicates any involvement by the State, other than expressing its immediate desires as a Member State of those organisations in consultation within the context of the UN or NATO, and that the Foundation et al.’s offer of proof through the examination of witnesses would be disregarded for that reason.
complains that the Court of Appeal’s opinion in para. 29.6 is legally incorrect, as the Court of Appeal exceeded the limits of the legal battle and/or wrongly added factual grounds on its own initiative and/or rendered a surprising decision, as the State exclusively invoked Article 59(1) DARIO and did not rely on the fact that the decision-making regarding the discontinuation of air support took place in accordance with NATO’s rules.
In the discussion of this part, I would put first and foremost that the Court of Appeal did not express an opinion here regarding the attribution of conduct of an organ of the State made available to an international organisation, but merely gave an opinion on attribution of responsibility for decisions of an international organisation itself to the State. These complaints are directed against the Court of Appeal’s opinion as to whether the State had effective control, within the meaning of Article 7 DARIO, of requests for air support or frustrating the same, which were already discussed and found to lack grounds above. Article 59 DARIO reads as follows:
“1. A State which directs and controls an international organization in the commission of an internationally wrongful act by the latter is internationally responsible for that act if:
(a) the State does so with knowledge of the circumstances of the internationally wrongful act; and
(b) the act would be internationally wrongful if committed by that State.
2. An act by a State member of an international organization done in accordance with the rules of the organization does not as such engage the international responsibility of that State under the terms of this draft article”.
Paras. 2 and 4 of the ILC’s Commentary on Article 59 DARIO37 assert that a distinction must be made between a Member State’s participation in the decision-making process based on the applicable rules and the direction and control based on which responsibility may be attributed to a State. Article 59(2) DARIO merely specifies that, in any case, no attribution may be made (on the ground that a State has exercised direction and control) if the situation involves conduct of a Member State that is accordance with the rules of the organisation.
No consideration need be given to whether the Court of appeal was permitted to base its finding in para. 29.6 on Article 59(2) DARIO, since the Court of Appeal also held that, leaving aside the question of whether the State’s acts were in accordance with NATO rules, the State did not exercise any direction and control over the UN or NATO. The Court of Appeal also found that Member States may provide a certain degree of political input without assuming responsibility for all or part of the decisions taken by the UN or NATO. The Court of Appeal thus manifestly found that the influence with the State exercised on the decisions taken by the UN and NATO did not result in the relevant decisions being implemented under the direction and control of the State. The Court of Appeal also held that that finding cannot be diminished by the pressure that NATO allegedly put on the Netherlands to permit air strikes and that the UN interrupted or discontinued close air support after a telephone conference between Voorhoeve and Akashi on the subject. Given the foregoing, the Foundation et al. have no interest in their complaint about the Court of Appeal’s finding regarding Article 59(2) DARIO, so the complaint fails.
complains that the Court of Appeal made no findings of fact regarding the procedural and other rules applicable within NATO and the UN, as a result of which the Court of Appeal could not verify whether the State’s involvement was in accordance with the rules of those organisations. relates to the offer of proof made by the Foundation et al., which the Court of Appeal ignored.
The complaints in parts 3.2 and 3.3 built on the complaint in part 3.1 and thus they must also share its fate.
Parts 4 and 5 (grounds for cross-appeal in cassation): direct effect of the Genocide Convention?
Uitvoeringswet Genocideverdrag);38 a statutory scheme in compliance with the obligation to prevent genocide was not created. That obligation is best fleshed out by the court; the court is able to determine for each individual act of genocide which measures would have been appropriate to prevent such acts. The obligation of preventing genocide included in Article I is unconditional in that sense, and can be sufficiently accurately applied by the court in each specific case, meaning according to the complaint that these provisions can serve as objective law.is directed against para. 34.4 of the contested judgment, where the Court of Appeal held that the obligation to prevent genocide from Article I of the Genocide Convention does not have direct effect. The part complains that this obligation, in light of the provisions of the Genocide Convention, does have direct effect. Article I of the Genocide Convention entails the unconditional obligation to prevent and punish genocide. Article II describes in detail the conduct that is considered genocide. Article V stipulates that the Contracting Parties are required to enact legislation and in particular to provide for effective penalties against those guilty of the acts listed in Articles II and III. At the time, in compliance with this obligation the Netherlands implemented the Genocide Convention Implementation Act (
Regarding the direct effect of treaties, I would note the following. The question of whether a treaty provision has direct effect within the meaning of Articles 93 and 94 of the Constitution must be determined by interpreting the treaty provisions involved based on Articles 31-33 of the Vienna Convention on the Law of Treaties (hereinafter: the Vienna Convention)39. If it cannot be concluded based on either the text or the legislative history of the provision that there is no direct effect, the content of the provision is decisive. The question is whether the provision is unconditional and sufficiently specific to be directly applied in the national system of law within the context in which it is being invoked to be applied as objective law40. Article 4 of the Vienna Convention stipulates that the Convention has no retroactive effect and exclusively applies to treaties that are adopted by states after the Vienna Convention takes effect. The interpretation rules of Articles 31-33 of the Vienna Convention, however, are generally considered codification of the applicable customary international law, meaning that these rules may be applied to older treaties.41 The Genocide Convention42 is such an older convention because it was adopted on 9 December 1948, and therefore amply before the Vienna Convention was adopted.
It cannot be concluded from either the text or the legislative history of the creation of the Genocide Convention that the Contracting States agreed that no direct effect may be assigned to Article I of the Genocide Convention. To answer the question of whether the obligation to prevent genocide from Article I of the Genocide Convention has direct effect, therefore, the content of that provision must be assessed. I would note in general that the obligations included in the Genocide Convention cannot be readily applied in national law without additional legislation. After all, Article V of the Genocide Convention provides as follows, citing the authentic English text:
“The Contracting Parties undertake to enact, in accordance with their respective Constitutions, the necessary legislation to give effect to the provisions of the present Convention and, in particular, to provide effective penalties for persons guilty of genocide or of any of the other acts enumerated in article III.”
From this it already follows that the provisions of the Genocide Convention have no direct effect.43 Articles II and II of the convention, which define genocide and make it punishable, do not contain any exact sanction or penalty. This general context is also relevant to the interpretation of Article I as such.
Article I of the Genocide Convention reads as follows in the authentic English text:
“The Contracting Parties confirm that genocide, whether committed in time of peace or in time of war, is a crime under international law which they undertake to prevent and to punish.”
In the Dutch translation:
“De Verdragsluitende Partijen stellen vast, dat genocide, ongeacht of het feit in vredes- dan wel in oorlogstijd wordt bedreven, een misdrijf is krachtens internationaal recht, welk misdrijf zij op zich nemen te voorkomen en te bestraffen.”
The obligation to make genocide punishable and the obligation to prevent genocide are two separate obligations. According to the International Court of Justice, the obligation to prevent genocide is not absorbed by the obligation to punish genocide.44 Unlike the obligation to punish genocide, which is fleshed out in the other articles of the Genocide Convention, the obligation to prevent genocide is not fleshed out, other than in Article VIII. This is why it is much more the case than with the obligation to punish genocide, which – in view of the above – has no direct effect, to clarify the scope of the obligation to prevent genocide that ensues from Article I of the Genocide Convention and the specific measures that this obligation demands.45
It follows from the foregoing that the Court of Appeal did not start from an incorrect interpretation of the law by ruling that the obligation to prevent genocide is not sufficiently specific for application as objective law. The complaint in part 4 therefore fails.
is directed against paras. 34.5, 50.1 and 51.6 of the judgment, and can be divided into three sub-parts. complains, briefly put, that the Court of Appeal started from an incorrect interpretation of the law in so far as paras. 50.1 and 51.6 might entail that the Genocide Convention not only has no direct effect, but also cannot have effect in the national system of law via application of national law.
The complaint starts from an incorrect reading of the judgment of the Court of Appeal and fails for that reason. After all, the Court of Appeal did not rule in para. 50.1 that the Genocide Convention cannot have effect in the national system of law, but merely ruled that it would not assess whether there was knowledge of genocide, as the conduct of Dutchbat could also be unlawful if the knowledge existed that the men to be evacuated ran a real risk of death or inhumane treatment.
complains that the opinion of the Court of Appeal is incorrect or insufficiently reasoned if the Court of Appeal intended to rule that an assessment based on Articles 2 and 3 ECHR and Articles 6 and 7 ICCPR would have the same consequences as an assessment based on the obligations from the Genocide Convention, which have effect through the standard of due care, whether or not assessment against the provisions of the ECHR and the ICCPR would more readily lead to a conclusion of wrongfulness. This is because the question of the effect of the obligations from the Genocide Convention is relevant not only in connection with the evacuation, but also in connection with the other acts of which Dutchbat is accused, according to this part.
In the considerations contested by this part, the Court of Appeal expressed that it is not necessary to assess whether knowledge of genocide existed with regard to the opinion that the State acted in breach of a statutory obligation and/or the obligation to do what is socially accepted according to unwritten law (cf. para. 39.2). If knowledge already exists of a real risk of genocide, this necessarily means that knowledge also existed of a real risk of death or inhumane treatment. This means that no case can be imagined in which there is knowledge of a real risk of genocide, but no knowledge of a real threat of death or inhumane treatment. Conversely, knowledge of a real danger of death or inhumane treatment can of course already exist without there being knowledge of a real risk of genocide. In order to determine whether the State acted unlawfully, the Court of Appeal could therefore suffice with assessing whether knowledge existed of a real danger of death or inhumane treatment. It is impossible to understand the grounds on which the legal consequences of the Court of Appeal’s opinion about unlawfulness might differ depending on whether the Court of Appeal assumed that knowledge existed of a real risk of genocide or assumed that knowledge existed of a real danger of death or inhumane treatment. The Genocide Convention contains no provisions in this regard. The complaint fails to that extent. As the Court of Appeal already rejected the accusations made by the Foundation et al. in respect of the other acts on the basis of other grounds, the complaint also fails in that regard.
complains that the Court of Appeal did not respond with sufficient reasons to ground of appeal 1 on the side of the State, where the State put forward that the Genocide Convention does not have effect in the national system of law. This complaint embroiders upon the previous parts and must share their fate.
Part 6 (ground for cross-appeal in cassation): assessment by the Court of Appeal of the unlawfulness of Dutchbat’s conduct
is directed against para. 39.2 of the contested decision, in which the Court of Appeal held that within the context of the assessment of the unlawfulness of Dutchbat’s conduct, the court also assesses whether Dutchbat could have reasonably decided and acted as it did, and that there is no ground for further restraint in that assessment. This part can be divided into three sub-parts.
complains that the Court of Appeal expressed an incorrect interpretation of the law in so far as it intended to rule that, in the event that a right has been infringed or an act has been performed that conflicts with a statutory obligation, whether Dutchbat was reasonably able to decide and act as it did must also be assessed. argues that the Court of Appeal started from an incorrect interpretation of the law in so far as it meant to say that there is no ground for “further restraint” to express that a different assessment, with more restraint, should be performed than the assessment of whether Dutchbat acted carefully, and that the conduct of Dutchbat should not be assessed in full. This holds particularly true, according to the complaint in , in so far as this involves military decisions, orders and instructions in which there is no margin of appreciation other than that which ensues from or is inherent in the interpretation of the decisions, orders and instructions.
The parties hold conflicting opinions regarding what may be concluded from the Supreme Court’s judgments in the [A] and [B] cases46. In para. 3.18.2 of the two judgments, the Supreme Court held that the Court of Appeal did not assess the conduct of Dutchbat, contrary to what the State argued in cassation, with hindsight, but always applied the standard of whether Dutchbat was reasonably able to decide and act as it did. However, the Supreme Court did not explicitly endorse that standard. In para. 3.18.3 the Supreme Court then ruled that there is no basis for the assessment with restraint advocated by the State – namely that the Court of Appeal actually only should have assessed whether Dutchbat could have acted as it did.
“3.18.3 In so far as the parts accuse the Court of Appeal of failing to appreciate the necessity of an assessment with restraint, they miss the mark, as no basis for such an assessment with restraint can be found in the international unwritten law, the ECHR or the ICCPR, or, incidentally, in Dutch national law.
The assessment with restraint advocated in these parts would entail that there would be virtually no margin of appreciation by the court of the consequences of the conduct of troops within the context of a peacekeeping mission – in this case: the conduct of Dutchbat and therefore of the State. Restraint that goes that far is unacceptable. This is not changed by the fact that the State expects this to have unfavourable consequences on the performance of peacekeeping missions by the United Nations and, more in particular, on the willingness of Member States to make troops available for such operations. This is because this should not impede the possibility of a judicial assessment after the fact of the conduct of the troops involved. In addition, the court should indeed take into account the fact that this concerns decisions taken under significant pressure in a situation of armed conflict, but the Court of Appeal did not fail to appreciate that.”
In the case now in cassation, the Court of Appeal rendered an opinion on the conduct of Dutchbat as such. This is because in essence, the Court of Appeal continually considered whether, given all of the circumstances of the case and given the knowledge that Dutchbat had or ought to have had, Dutchbat could have been reasonably expected to do more or otherwise in respect of the protection of the Bosnian Serbs (cf. paras. 53.8, 54.3e, 60.3 and 60.4, 61.4-61.6, and 63.2-63.7 of the contested judgment). The Court of Appeal therefore did not assess Dutchbat’s conduct with restraint and in para. 39.2, it did not start from an incorrect interpretation of the law. This means that the complaints in part 6 fail.
Part 7 (ground for cross-appeal in cassation): Dutchbat’s knowledge of the danger of genocide and violation of human rights
is directed against various opinions of the contested judgment, mentioned at 7.0, and can be divided into five sub-parts. The part concerns the Court of Appeal’s opinion regarding the question of what knowledge Dutchbat had of the danger of genocide or of violation of fundamental human rights.
contain the essence of the complaint. These parts argue that in so far as the Court of Appeal assumed that this merely concerns facts and circumstances of which Dutchbat had actual subjective knowledge (whether or not on the basis of attributing the knowledge of certain individuals to the organization), this opinion is based on an incorrect interpretation of the law, because what Dutchbat reasonably ought to have understood regarding the real risk of fundamental human rights being violated is also relevant.
These complaints fail, as they are based on an incorrect reading of the judgment. This is because in various paragraphs, the Court of Appeal refers to what Dutchbat “reasonably ought to have understood” (para. 46), “ought to have known” (para. 52.1) and “knew or ought to have known” (paras. 61.5 and 61.8). The Court of Appeal therefore did not limit its assessment to what Dutchbat actually subjectively knew, but also assessed what Dutchbat ought to have understood, given all of the circumstances of the case, regarding the existence of a real risk of fundamental human rights being violated.
is directed against the Court of Appeal’s opinion in para. 50.2 that those who had the authority to take any decision regarding the acts or omissions of Dutchbat must have had the asserted knowledge. The part is also directed against the Court of Appeal’s opinion in para. 50.4 that the “leadership” of Dutchbat should be understood to include all those who could take a relevant decision regarding the men, and therefore also the “commanders giving orders in the mini safe area, including Franken and Karremans”. The part complains that this opinion starts from an incorrect interpretation of the law or is incomprehensible because in order to attribute knowledge to Dutchbat, in any event the leadership of Dutchbat must be understood to include the company commanders, including Matthijssen and Groen, and their deputies, as well as the battalion’s warrant officer. Furthermore, the knowledge of those who specifically went out to investigate and collect information that could be relevant in the taking of decisions in the matter of the evacuation, and of those who can be deemed by virtue of that order to have shared their knowledge with the Dutchbat leadership, should also be attributed to Dutchbat.
The part puts to discussion the question of whether the knowledge of officers of a legal entity – internal knowledge47 – can be attributed to that legal entity. In this, what matters is whether this knowledge is deemed according to accepted standards to be knowledge of the legal entity48. In answering the question of whether the basis of the attribution should be broad or less broad, relevant are the circumstances of the case as well as the nature and substance of the legal relationship to be assessed.49
The Court of Appeal reasoned its contested opinion by referring to the given that individual troop members could not take decisions and to the nature and circumstances of the case, to wit a brief, chaotic period of two days during which thousands of people were awaiting evacuation under deplorable circumstances and in which a personnel shortage occurred shortly after Dutchbat was forced to abandon its mission. In view of the foregoing, the Court of Appeal was entitled to draw the conclusion from these circumstances that the knowledge of individual troop members, including the individual troop members mentioned in the part who went out to investigate, cannot, according to accepted standards, be unconditionally deemed knowledge of the leadership of Dutchbat and, by extension, as knowledge of the State. This is why the Court of Appeal’s opinion does not reflect an incorrect interpretation of the law and is not incomprehensible, as a result of which this part fails.
also complains that the Court of Appeal’s opinion was insufficiently reasoned in so far as the Court of Appeal intended to rule that only Karremans and Franken, and not the company commanders and their deputies, were part of the “leadership” of Dutchbat.
This part fails, as it is based on an incorrect reading of the contested judgment. This is because the Court of Appeal considered not only Karremans’ and Franken’s knowledge, but also the knowledge of Van Duijn (the platoon commander)50. The Court of Appeal therefore did not fail to appreciate that people other than the battalion commander, Karremans, and the deputy battalion commander, Franken, are deemed to be included in the “leadership” of Dutchbat, such as the company commanders.
embroiders upon the previous sub-parts and therefore also fails.
Part 8 (ground for cross-appeal in cassation): Causal connection between Dutchbat’s conduct and the damage suffered; chance percentage
is directed against para. 64.1 et seq., where the Court of Appeal ruled on the causal connection between the conduct of Dutchbat and the damage suffered by the claimants. This part can be divided into five sub-parts.
argues that in the event that one or more of the complaints from the preceding parts is successful, the paragraphs mentioned in this part regarding the causal connection cannot be upheld. As the preceding parts fail, part 8.1 also fails.
prima facie that this proof has been provided by the Foundation et al. The Court of Appeal wrongly failed to do so, or at least did not provide sufficient reasons for its implicit opinion in this regard.complains that in view of the nature of the damage and the circumstance that there is uncertainty as to whether the fate of the men is attributable to unlawful conduct by the State, there is every reason to accommodate the Foundation et al. in the proof of the causal connection and to rule
The ground for cross-appeal in cassation argues that the Court of Appeal should have proceeded to rule “prima facie” on the proof in respect of the causal connection. Giving an opinion “prima facie” does not mean that this might be an opinion based on a first glance at the evidence. 51 The Court of Appeal held in para. 8.2 that it must be assumed that there are sufficient points of reference to presume that the relevant facts have been substantiated. The Court of Appeal subsequently held in para. 64.1 et seq. on the basis of a full evaluation of all of the available evidence that a causal connection – in the sense of a condicio sine qua nonconnection – is lacking between the unlawful conduct of Dutchbat and the fate of the men who were outside of the compound in the mini safe area on 13 July 1995. Again in paras. 66.2-68, the Court of Appeal evidently ruled on the basis of a full evaluation of all of the available evidence that although there is no causal connection between Dutchbat’s unlawful conduct and the actual fate of the men who were inside the compound on 13 July 1995, there is such a connection between the conduct of Dutchbat and the damage caused by refusing to give those men a chance of survival.52 Matters thus standing, it is impossible to imagine how the Court of Appeal’s final opinion could have been different if it had given a prima facie opinion regarding proof at any time in the proceedings and followed proof to the contrary based on the available evidence. The complaint therefore fails.
That, in cases like the present, a deviation from or reduction of the division of the burden of proof may be appropriate in general or in principle, as argued in parts 18.3.3-18.3.7 of the Foundation et al.’s written explanation, remains very much to be seen, in light of the opinion of the International Court of Justice in the case of the Bosnian Genocide:
“461. The Court has found that the authorities of the Respondent could not have been unaware of the grave risk of genocide once the VRS forces had decided to take possession of the Srebrenica enclave, and that in view of its influence over the events, the Respondent must be held to have had the means of action by which it could seek to prevent genocide, and to have manifestly refrained from employing them (paragraph 438). To that extent therefore it failed to comply with its obligation of prevention under the Convention. The obligation to prevent the commission of the crime of genocide is imposed by the Genocide Convention on any State party which, in a given situation, has it in its power to contribute to restraining in any degree the commission of genocide. To make this finding, the Court did not have to decide whether the acts of genocide committed at Srebrenica would have occurred anyway even if the Respondent had done as it should have and employed the means available to it. This is because, as explained above, the obligation to prevent genocide places a State under a duty to act which is not dependent on the certainty that the action to be taken will succeed in preventing the commission of acts of genocide, or even on the likelihood of that outcome. It therefore does not follow from the Court’s reasoning above in finding a violation by the Respondent of its obligation of prevention that the atrocious suffering caused by the genocide committed at Srebrenica would not have occurred had the violation not taken place.
462. The Court cannot however leave it at that. Since it now has to rule on the claim for reparation, it must ascertain whether, and to what extent, the injury asserted by the Applicant is the consequence of wrongful conduct by the Respondent with the consequence that the Respondent should be required to make reparation for it, in accordance with the principle of customary international law stated above. In this context, the question just mentioned, whether the genocide at Srebrenica would have taken place even if the Respondent had attempted to prevent it by employing all means in its possession, becomes directly relevant, for the definition of the extent of the obligation of reparation borne by the Respondent as a result of its wrongful conduct. The question is whether there is a sufficiently direct and certain causal nexus between the wrongful act, the Respondent’s breach of the obligation to prevent genocide, and the injury suffered by the Applicant, consisting of all damage of any type, material or moral, caused by the acts of genocide. Such a nexus could be considered established only if the Court were able to conclude from the case as a whole and with a sufficient degree of certainty that the genocide at Srebrenica would in fact have been averted if the Respondent had acted in compliance with its legal obligations. (…)”. (emphasis added, Advocate General)
It is clear from the judgment of the International Court of Justice that the presence of the causal link must always be established with sufficient certainty .
complains that the Court of Appeal failed to recognise that it is sufficient for a referral to proceedings for the determination of damages that the possibility of damage is plausible. The Court of Appeal therefore should have refrained from giving a further opinion on the plausibility of the causal link after it had established that there was a real chance that the men would have survived, and should have left that further assessment to the court hearing the proceedings for the determination of damages, according to this part of the ground for cassation.
It is important to note that, in order to refer the matter to proceedings for the determination of damages, it must be established that there is not a complete absence of a causal link, or at least that the assertion that there is no causal link at all must have been rejected by the court in the main proceedings.53 In addition, the court hearing the main proceedings can also express an opinion on the merits about the causal link. In its judgment of 27 October 2017, the Supreme Court considered as follows in that respect:
“3.4.3 Pursuant to Article 612 DCCP, the court that orders payment of damages will, in principle, assess the damage in its judgment as far as it is possible for it to do so, even if only compensation to be determined in separate follow-up proceedings has been claimed but sufficient statements have been made and proven to be able to order payment of a particular amount. This principle means that the court, to the extent that it is possible in light of the debate between the parties and with due observance of the principle of hearing both sides of the argument, can already decide in the main proceedings on points that divide the parties, even if it concerns points in dispute that could by themselves be brought up for further discussion in the proceedings for the determination of damages, such as matters like causal link and own fault (cf. Supreme Court 16 April 2010, ECLI:NL:HR:2010:BL2229, NJ 2010/229, para. 3.5.4)”.54
The Court of Appeal could therefore, without violation of any rule of law, make a statement on the merits about the plausibility of the existence of a causal link between the actions of Dutchbat and the lost opportunity damage suffered by the Foundation et al., at least as far as possible in light of the debate between the parties and with due observance of the principle of hearing both sides. The complaint therefore fails.
complains that the opinion of the Court of Appeal that the men’s chances of survival should be estimated at 30% is an unacceptable surprise decision, as this was not sufficiently debated between the parties.
The complaint fails. The court record of the oral arguments on appeal (pages 5-6) shows that further to questions asked by the Court of Appeal a debate took place between the parties on the question of what would have happened if the more than 300 men had stayed in the compound and whether they would then have had a reasonable chance to survive. On behalf of the Foundation et al., it was argued with substantiation that the men would have had a reasonable chance if they had stayed within the compound. On behalf of the State, it was argued with substantiation that the fate of the men would not have been any different in that case. In a letter dated 20 July 2017,55 the Foundation et al. requested that the court record be amended in such a way that the percentage of 50% – which according to the court record was linked by the Foundation to the reasonable chance of the men within the compound – was not mentioned in relation to those men, but only in relation to the group of men who fled through the woods. Whatever the case may be, given the course of the party debate it cannot be said that the parties need not have expected that the Court of Appeal, based on the assertions of the parties and the submitted documents, would weigh up the good and bad chances of the men within the compound and would determine the chance of the men’s survival as it did. In fact, it is clear from the letter of 20 July 2017 that the Foundation et al. took this into account.
complains that the Court of Appeal failed to recognise in para. 68, with regard to the claims brought by the Foundation, that the causal link need not be assessed in these proceedings. Specifically by establishing a 30% probability that the men would have escaped the inhumane treatment and execution by the Bosnian Serbs, the Court of Appeal exhibited an incorrect interpretation of the law, as the Foundation is not authorised to bring a claim for monetary compensation.
The opinion of the Court of Appeal contested by this part of the ground for cassation can only relate to the collective claim brought by the Foundation. After all, the claims of the other claimants were rejected because their husbands were not evacuated from the compound on 13 July 1995. The collective claim filed by the Foundation can only relate to the claim for a judicial declaration that the State acted wrongfully towards the surviving dependants whose interests are represented by the Foundation. After all, the claim for compensation of the damage suffered by the claimants, to be determined by the court in follow-up proceedings, concerns the claim of the individual claimants themselves,56 while pursuant to Article 3:305(3) DCC the Foundation is not authorised to institute legal proceedings aimed at obtaining monetary compensation. In the operative part of the contested judgment, the Court of Appeal issued a judicial declaration entailing that the State acted wrongfully “by not giving the male refugees who were inside the compound on 13 July 1995 the choice of staying behind in the compound, and thus denying them the 30% chance of not being exposed to the inhumane treatment and executions by the Bosnian Serbs”. In doing so, the Court of Appeal seems to have issued a slightly different judicial declaration than the declaration that was sought, as this concerns not only the wrongfulness of Dutchbat’s conduct but also the damage caused by it. However, the ground for cassation does not complain about this. The Court of Appeal has therefore apparently also expressed its opinion on the probability percentage in the context of the assessment of the claim for the relevant judicial declaration, rather than just for the referral to the proceedings for the determination of damages. 57 Consequently, the complaint fails.
4.100 I would like to make the following comments about the assessment of the causal link and the extent of the damage in the context of a collective action on the basis of Article 3:305a DCC. Pursuant to Article 3:305a(1) DCC, a foundation or an association with full legal capacity may institute legal proceedings for the protection of similar interests of other persons, in so far as it represents these interests pursuant to its articles. Pursuant to Article 3:305a(1) DCC, for example, a judicial declaration can be sought entailing that the defendant has acted wrongfully towards the persons whose interests are represented by the foundation or association. The third paragraph of Article 3:305a DCC provides that no monetary compensation can be claimed in a collective action. The option of instituting a collective action for damages is ruled out, since the legislator believes that, in dealing with such a claim, questions would be raised that can be answered only for each individual injured party on the basis of the specific circumstances relating to that party, such as questions of causality, own fault and the extent of the damage.58 Therefore, in collective proceedings, the court will (in principle) not be able to rule on claims which essentially serve to establish the extent of the obligation to pay compensation to each of the individual injured parties. In the context of a collective action, only those questions that are “generalisable”, i.e. questions that lend themselves to a collective hearing, can be addressed. I quote the following in this regard from the Supreme Court’s judgment on Vie d’Or of 13 October 2006:
“8.1.3 In essence, the complaints of these parts are based on the position that the present claim for a judicial declaration that the Insurance Supervisory Authority, the actuary and the auditors are jointly and severally liable to compensate the damage suffered by the former policyholders in the bankruptcy of Vie d’Or is one of the claims that in connection with the similar interests of the policyholders to be bundled can be brought in proceedings pursuant to Article 3:305a DCC and that this joint and several liability is not barred by the fact that these claims, pursuant to paragraph 3 of that provision, cannot seek monetary compensation.” That position is incorrect.
The Court of Appeal ruled – not incomprehensibly – that the purpose of the claim in question is essentially to determine the extent of the liability for compensation to each of the individual policyholders. As this determination cannot be made without considering the extent to which, depending on the special circumstances of the case, the occurrence of that individual damage can be attributed to the conduct of the Insurance Supervisory Authority, the actuary and the auditors and the extent to which the circumstances that contributed to the damage can be attributed to the former and possibly to the individual injured parties, the purport of Article 3:305a DCC precludes the granting of that claim. As the Court of Appeal has not incomprehensibly considered, the interests which the legal claim thus aims to serve cannot be generalised to such an extent that they cannot be counted among the similar interests to which Article 3:305a DCC refers. This is not changed by the fact that the Court of Appeal, in particular with regard to the causal link between the wrongful conduct of the Insurance Supervisory Authority, the actuary and the auditors and the total damage and the time at which this total damage must be calculated, has rendered decisions that in themselves concern ‘generalisable questions of causality and liability’.” 59 (emphasis added, A-G)
4.101 In the above judgment, the Supreme Court left open the possibility that causality issues are also assessed in the context of a collective action, provided that these issues can be sufficiently generalised. It therefore depends on the circumstances of the case.60 Generally speaking, it cannot be said that, for example, the condicio sine qua non connection can be assessed in proceedings further to a collective action, while other causality problems or questions about the damage cannot be assessed.61 It is always up to the court to decide, in the specific case, which issues can be sufficiently generalised and which cannot.62 I would like to point out that the bill on the settlement of large-scale damage in collective actions, currently still pending before the States General, provides for the possibility of claiming compensation in a collective action. The Explanatory Memorandum to this bill states, as a disadvantage of the existing provision of Article 3:305a DCC, that some questions must be assessed separately for each individual victim, including questions of causality, prescription and own fault.63
4.102 Under current law, it is in principle not possible to rule on causality and the extent of the damage in the context of collective proceedings on the basis of Article 3:305a DCC.64 In the cases in which it is possible to abstract from and therefore “generalise” the special circumstances of individual cases to form an opinion on the causal link and the damage, as in cases in which groups of persons were in almost all respects in the same position with regard to the person who acted wrongfully, the court can comment on this in collective proceedings. In the contested judgment, the Court of Appeal apparently ruled that in the present case it was possible, with a sufficient degree of “generalisability”, to comment on the causal link between the actions of Dutchbat and the opportunity that was lost, as well as on the extent of the lost opportunity. The Court of Appeal was therefore able to give an opinion on those aspects of the case without violating any rule of law. The complaint of part 8.5 therefore fails.
4.103 The conclusion is that the ground for cross-appeal in cassation fails.
5 Discussion of the principal ground for cassation
The principal ground for cassation consists of five parts, of which parts 1 through 4 are divided into various sub-parts. Parts 1 through 3 concern Dutchbat command’s knowledge of the fate that awaited the men (part 1), the Court of Appeal’s opinion on Dutchbat’s actions at the “sluice” during the evacuation on 13 July 1995 (part 2) and Dutchbat’s actions with regard to the men who were at the compound on 13 July 1995 (part 3).
The parts contain, among other things, complaints about the assessment standard applied by the Court of Appeal in respect of Dutchbat’s actions. The Court of Appeal found that the question of whether there was question of wrongful conduct within the meaning of Article 6:162 DCC should be assessed under Dutch law. This finding has not been contested in cassation. In the interpretation of what is generally accepted according to unwritten law, the Court of Appeal apparently took into account the international standards on human rights and international law, such as the obligation to prevent or provide protection against the violation of fundamental human rights by third parties. Another finding of the Court of Appeal left uncontested in cassation is that the State had jurisdiction within the compound within the meaning of Article 1 ECHR and Article 2(1) ICCPR, so that the State in that respect also had a direct duty to observe its (positive) obligations under Articles 2 and 3 ECHR and Articles 6 and 7 ICCPR vis-à-vis those present at the compound.
Before discussing the complaints of the principal ground for cassation, I will devote some attention to the State’s positive obligation to prevent genocide and violations of fundamental rights by third parties. I will then discuss the question of which human rights obligations apply to states within the context of war situations and/or military operations.
The State’s obligation to prevent genocide
The Genocide Convention and the question of whether that convention has a direct effect have already been addressed in the discussion of parts 4 and 5 of the ground for cross-appeal. I also quote the text of Article I of the Genocide Convention in number 4.69 of my opinion, so that it is sufficient to refer to that quote. Pursuant to Article I of the Genocide Convention, states are obliged to prevent and punish genocide. This means, first and foremost, that states are obliged to prevent genocide from taking place on their own territory. It follows from the judgment of the International Court of Justice in the Bosnian Genocide matter (par. 430) that Article I of the Genocide Convention also obliges Contracting Parties to prevent genocide from taking place outside their territory.65 The Court of Justice held that this is a best-efforts obligation, not a result obligation:
“430. (…) it is clear that the obligation in question is one of conduct and not one of result, in the sense that a State cannot be under an obligation to succeed, whatever the circumstances, in preventing the commission of genocide: the obligation of States parties is rather to employ all means reasonably available to them, so as to prevent genocide so far as possible. A State does not incur responsibility simply because the desired result is not achieved; responsibility is however incurred if the State manifestly failed to take all measures to prevent genocide which were within its power, and which might have contributed to preventing the genocide. In this area the notion of “due diligence”, which calls for an assessment in concreto, is of critical importance. Various parameters operate when assessing whether a State has duly discharged the obligation concerned. The first, which varies greatly from one State to another, is clearly the capacity to influence effectively the action of persons likely to commit, or already committing, genocide. This capacity itself depends, among other things, on the geographical distance of the State concerned from the scene of the events, and on the strength of the political links, as well as links of all other kinds, between the authorities of that State and the main actors in the events. The State’s capacity to influence must also be assessed by legal criteria, since it is clear that every State may only act within the limits permitted by international law; seen thus, a State’s capacity to influence may vary depending on its particular legal position vis-à-vis the situations and persons facing the danger, or the reality, of genocide. On the other hand, it is irrelevant whether the State whose responsibility is in issue claims, or even proves, that even if it had employed all means reasonably at its disposal, they would not have sufficed to prevent the commission of genocide. As well as being generally difficult to prove, this is irrelevant to the breach of the obligation of conduct in question, the more so since the possibility remains that the combined efforts of several States, each complying with its obligation to prevent, might have achieved the result — averting the commission of genocide — which the efforts of only one State were insufficient to produce”.
It is therefore not necessary to establish that the measures omitted would have actually prevented the genocide in order to find that a state has failed in its obligations under Article I of the Genocide Convention. According to the Court of Justice, a state must take such measures as are reasonably available to it to prevent genocide so far as possible. In this respect, the Court of Justice applies the concept of due diligence. This is a well-known standard in public international law for assessing the wrongfulness of state actions. Due diligence refers to the existence of a best-efforts obligation: taking all reasonably possible measures in order to prevent a certain harmful effect as much as possible.66 When that obligation arises and how it should be implemented depends on various factors.
It follows from the consideration of the International Court of Justice cited above that the first determining factor is the extent to which the state can effectively influence the actions of those committing or threatening to commit genocide. This also depends on various factors, including the geographical proximity of the state to the place where the atrocities take place or threaten to take place, the strength of political and economic ties and all other ties of the state with those who have acted. The stronger those ties are, the more influence the state can exert and the wider the scope of the obligation to prevent genocide becomes.67 The fact that a state is present with peacekeeping forces usually means that the state can influence the actions of those committing or threatening to commit genocide to some degree.68 The extent of that influence depends on the mandate – as states are not expected to act in breach of international law69 – and the troop strength, the presence of further support, etc. It is clear, however, that the influence that a state can exert through the presence of peacekeeping forces is of a completely different nature from the influence that a state can exert through political or economic ties or through geographical proximity. This should be taken into account when assessing what can reasonably be expected of a state.
The International Court of Justice furthermore considered in the Bosnian Genocide judgment (para. 431) that the obligation to prevent genocide can only be violated when the genocide has actually taken place, but that this does not mean that the obligation only takes effect once the genocide has begun. After all, that would render the obligation to prevent that genocide meaningless.70 The Court of Justice held that the obligation to act arises at the moment that the State becomes aware, or should normally have become aware, of a serious risk of genocide. Although there was no preconceived plan in respect of the Srebrenica genocide, Serbia, in the person of (then) President Milošević, given also his contact with the Bosnian-Serb army commander Mladić, was aware of what was about to happen. In par. 438, the Court of Justice held that there was a serious risk of genocide in Srebrenica:
“The FRY leadership, and President Milošević above all, were fully aware of the climate of deep-seated hatred which reigned between the Bosnian Serbs and the Muslims in the Srebrenica region. As the Court has noted in paragraph 423 above, it has not been shown that the decision to eliminate physically the whole of the adult male population of the Muslim community of Srebrenica was brought to the attention of the Belgrade authorities. Nevertheless, given all the international concern about what looked likely to happen at Srebrenica, given Milošević’s own observations to Mladić, which made it clear that the dangers were known and that these dangers seemed to be of an order that could suggest intent to commit genocide, unless brought under control, it must have been clear that there was a serious risk of genocide in Srebrenica. Yet the Respondent has not shown that it took any initiative to prevent what happened, or any action on its part to avert the atrocities which were committed. It must therefore be concluded that the organs of the Respondent did nothing to prevent the Srebrenica massacres, claiming that they were powerless to do so, which hardly tallies with their known influence over the VRS. As indicated above, for a State to be held responsible for breaching its obligation of prevention, it does not need to be proven that the State concerned definitely had the power to prevent the genocide; it is sufficient that it had the means to do so and that it manifestly refrained from using them”.
The positive obligation of the State to prevent violations of Articles 2 and 3 of the ECHR by third parties
It is established ECtHR case law that states are obliged under the ECHR to prevent the exposure of persons under their jurisdiction to the risk of violations of their ECHR rights in other countries. For example, the ECHR prohibits the extradition or deportation of persons to countries where they would be subjected to inhuman or degrading treatment (the principle of non-refoulement).71 Nor may persons be extradited or surrendered if they would run the risk of being sentenced to death, as the 13th Protocol to the ECHR prohibits the death penalty under all circumstances.72
States are obliged to ensure prior to extradition or surrender that the rights of those involved will be respected. The state must establish whether there is a real risk that this will not be the case. In Soering/United Kingdom, the ECtHR ruled in par. 91:
“In sum, the decision by a Contracting State to extradite a fugitive may give rise to an issue under Art. 3, and hence engage the responsibility of that State under the Convention, where substantial grounds have been shown for believing that the person concerned, if extradited, faces a real risk of being subjected to torture or to inhuman or degrading treatment or punishment in the requesting country”. 73
In the Osman/United Kingdom matter, the ECtHR formulated the positive obligation of the contracting states under Article 2 ECHR in situations where the life of a certain individual is threatened by a third party and held that this obligation should be interpreted with some restraint to avoid imposing an impossible and disproportionate burden on the authorities:
“115. (…) Article 2 of the Convention may also imply in certain well-defined circumstances a positive obligation on the authorities to take preventive operational measures to protect an individual whose life is at risk from the criminal acts of another individual. (…)
116. For the Court, and bearing in mind the difficulties involved in policing modern societies, the unpredictability of human conduct and the operational choices which must be made in terms of priorities and resources, such an obligation must be interpreted in a way which does not impose an impossible or disproportionate burden on the authorities. Accordingly, not every claimed risk to life can entail for the authorities a Convention requirement to take operational measures to prevent that risk from materialising. (…) In the opinion of the Court where there is an allegation that the authorities have violated their positive obligation to protect the right to life in the context of their above-mentioned duty to prevent and suppress offences against the person (…), it must be established to its satisfaction that the authorities knew or ought to have known at the time of the existence of a real and immediate risk to the life of an identified individual or individuals from the criminal acts of a third party and that they failed to take measures within the scope of their powers which, judged reasonably, might have been expected to alleviate that risk. (…) For the Court, and having regard to the nature of the right protected by Article 2, a right fundamental in the scheme of the Convention, it is sufficient for an applicant to show that the authorities did not do all that could be reasonably expected of them to avoid a real and immediate risk to life of which they have or ought to have knowledge. This is a question which can only be answered in the light of all the circumstances of any particular case”.74
In the Kilic/Turkey75 and Mahmut Kaya/Turkey76 cases, the ECtHR applied the Osman/United Kingdom standard to the respective murders of a journalist and a physician in an area within Turkey where there was a conflict between the Turkish army and the PKK and other groups. In both cases, the ECtHR ruled that the Turkish authorities had violated their positive obligation under Article 2 ECHR by not taking the reasonable measures available to them to protect the lives of the victims. In these cases, however, although it could not be established that any official representative of Turkey had had a hand in the murders, there were strong indications that the murders had been committed with the agreement of the Turkish authorities77 or that the real danger came from groups operating with the agreement of elements within the Turkish security services.78
The ECtHR has repeatedly held that the obligation to take preventive measures in order to protect the lives of persons not only extends to one or more specific persons at risk, but also pertains to the more general protection of society. This also applies in the Tagayeva/Russia case.79 The case concerned the taking of more than 1,000 people, including 800 school children, hostage by Chechen separatists in Beslan and the subsequent violent liberation by the Russian authorities in which over 300 people were killed. The ECtHR (par. 490) held that the Russian authorities had specific information on the size and nature, time and location of the Chechen attack, so that the authorities had sufficient control over the situation at least in the days preceding the attack. Despite the fact that the authorities had sufficient opportunity to prepare, 30 terrorists were able to get together, train and move to the school without being hindered by any preventive security measures (par. 491). The Court of Justice then considered:
“492. The Court reiterates that in the preparation of responses to unlawful and dangerous acts in highly volatile circumstances, competent law-enforcement services such as the police must be afforded a degree of discretion in taking operational decisions. Such decisions are almost always complicated, (…). In the face of an urgent need to avert serious adverse consequences, whether the authorities choose to use a passive approach of ensuring security of the potential targets or more active intervention to disrupt the menace, is a question of tactical choice. However, such measures should be able, when judged reasonably, to prevent or minimise the known risk. With regard to the above arguments, the Court finds that in the case at issue the Russian authorities failed to take such measures”.
In the earlier case Finogenov/Russia80case, the ECtHR arrive at a comparable opinion regarding the preparations for a military operation conducted by the Russian authorities. That case concerned the occupation by Chechen separatists of the Dubrovka theatre in Moscow. The occupation put to an end by the Russian authorities by storming the building after injecting an unknown intoxicating gas. The gas caused the death of 125 hostages. The Court of Human Rights held as follows:
“265 (…), the Court stresses that its role is not to establish the individual liability of those involved in the planning and coordination of the rescue operation (…). The Court is called upon to decide whether the State as a whole complied with its international obligations under the Convention, namely its obligation to “take all feasible precautions in the choice of means and methods of a security operation mounted against an opposing group with a view of avoiding and, in any event, minimising, incidental loss of civilian life” (…).
266. The Court acknowledges that in such situations some measure of disorder is unavoidable. It also recognises the need to keep certain aspects of security operations secret. However, in the circumstances the rescue operation of 26 October 2002 was not sufficiently prepared, in particular because of the inadequate information exchange between various services, the belated start of the evacuation, limited on-the-field coordination of various services, lack of appropriate medical treatment and equipment on the spot, and inadequate logistics. The Court concludes that the State breached its positive obligations under Article 2 of the Convention”.81
I will be returning to the Fingenov/Russia case, as that decision is relevant to the assessment of the State’s obligations on the basis of Article 2 ECHR in a military context in general and therefore within the context of the assessment of the positive obligations on the basis of Article 2 ECHR. In Finogenov/Russian, incidentally, the ECtHR noted that in the use of violence, which was at issue in that case, it is difficult to distinguish between a State’s negative and positive obligations:
“208. When lethal force is used within a “policing operation” by the authorities it is difficult to separate the State’s negative obligations under the Convention from its positive obligations. In such cases the Court will normally examine whether the police operation was planned and controlled by the authorities so as to minimize, to the greatest extent possible, recourse to lethal force and human losses, and whether all feasible precautions in the choice of means and methods of a security operation were taken (…)”.
The Court appears to include the planning aspects of the operation among the positive obligations under Article 2 ECHR and to consider the use of violence a negative obligation under Article 2 ECHR. This latter aspect will be discussed below.
Human rights obligations of states in war situations and/or in military operations
Humanitarian international law and international law in respect of human rights were traditionally considered separate areas of law that are supplementary but that cannot apply simultaneously. Now, it is generally accepted in the literature that the two areas of law have converged to a certain extent.82 In its Advisory Opinion on the Israeli Wall of 9 July 2004, the International Court of Justice held:
“106. More generally the Court considers that the protection offered by human rights conventions does not cease in case of armed conflict, save through the effect of provisions for derogation of the kind to be found in Article 4 of the International Covenant on Civil and Political Rights. As regards the relationship between international humanitarian law and human rights, there are thus three possible situations: some rights may be exclusively matters of international humanitarian law; others may be exclusively matters of human rights law; yet others may be matters of both these branches of law”.83
It can also be concluded from the case law of the ECtHR that the two areas of law have converged to a certain extent. In the Varnava/Turkey case, the ECtHR ruled that Article 2 ECHR must be interpreted
“in so far as possible in light of the general principles of international law, including the rules of international humanitarian law which play an indispensable and universally accepted role in mitigating the savagery and inhumanity of armed conflict (…)”.84
The ECtHR continues:
“The Court therefore concurs with the reasoning of the Chamber in holding that in a zone of international conflict Contracting States are under an obligation to protect the lives of those not, or no longer, engaged in hostilities”.
In the case Hassan/United Kingdom concerning the detention of an Iraqi citizen by British troops during the invasion of Iraq in 2003, the ECtHR also held that humanitarian international law and the law in respect of human rights have converged.85 The ECtHR has nevertheless accepted that in war situations, in which the State involved has not invoked the power laid down in Article 15 ECHR to derogate from treaty obligations, those treaty obligations (at least those under Article 5) can be moderated to a certain degree. The Court of Human Rights held as follows:
“103. (…) the Court accepts the Government’s argument that the lack of a formal derogation under Article 15 does not prevent the Court from taking account of the context and the provisions of international humanitarian law when interpreting and applying Article 5 in this case.
104. Nonetheless, and consistently with the case-law of the International Court of Justice, the Court considers that, even in situations of international armed conflict, the safeguards under the Convention continue to apply, albeit interpreted against the background of the provisions of international humanitarian law. By reason of the co-existence of the safeguards provided by international humanitarian law and by the Convention in time of armed conflict, the grounds of permitted deprivation of liberty set out in subparagraphs (a) to (f) of that provision should be accommodated, as far as possible, with the taking of prisoners of war and the detention of civilians who pose a risk to security under the Third and Fourth Geneva Conventions. The Court is mindful of the fact that internment in peacetime does not fall within the scheme of deprivation of liberty governed by Article 5 of the Convention without the exercise of the power of derogation under Article 15 (…). It can only be in cases of international armed conflict, where the taking of prisoners of war and the detention of civilians who pose a threat to security are accepted features of international humanitarian law, that Article 5 could be interpreted as permitting the exercise of such broad powers”.
The ECtHR is also willing to weigh the obligations under Article 2 ECHR differently within the context of the execution of military operations than would otherwise be the case.86 In the case Finogenov/Russia cited above, the ECtHR held that derogation from the standard for the use of lethal force, to wit the standard of absolute necessity, is possible in cases in which certain aspects of the situation extend far beyond the expertise of the Court and in which the authorities must act under extensive time pressure but have only minimal control of the situation. I quote the Court:
“210. As a rule, any use of lethal force must be no more than “absolutely necessary” for the achievement of one or more of the purposes set out in paragraph 2 (a), (b) and (c) of Article 2 of the Convention. This term indicates that a stricter and more compelling test of necessity must be employed by the Court, if compared with that normally applicable when determining whether State action is “necessary in a democratic society” under paragraphs 2 of Articles 8 to 11 of the Convention. Consequently, the force used must be strictly proportionate to the achievement of the permitted aims (see McCann and Others, cited above, §§ 148-49; see also Gül v. Turkey, no. 22676/93, §§ 77 and 78, 14 December 2000).
211. That being said, the Court may occasionally depart from that rigorous standard of “absolute necessity”. As the cases of Osman, Makaratzis, and Maiorano and Others (all cited above) show, its application may be simply impossible where certain aspects of the situation lie far beyond the Court’s expertise and where the authorities had to act under tremendous time pressure and where their control of the situation was minimal”.
The ECtHR went on in Finognov/Russia to rule that, in that case, the lives of hundreds of hostages were at stake and the authorities were forced to deal with trained and highly dedicated terrorists, while the hostage taking itself came as a surprise, and the authorities had no control of the building in which the hostages were being held. In such situations, the Court is willing to give the authorities a certain margin of appreciation:
“213. Although hostage taking has, sadly, been a widespread phenomenon in recent years, the magnitude of the crisis of 23-26 October 2002 exceeded everything known before and made that situation truly exceptional. The lives of several hundred hostages were at stake, the terrorists were heavily armed, well-trained and devoted to their cause and, with regard to the military aspect of the storming, no specific preliminary measures could have been taken. The hostage-taking came as a surprise for the authorities (…), so the military preparations for the storming had to be made very quickly and in full secrecy. It should be noted that the authorities were not in control of the situation inside the building. In such a situation the Court accepts that difficult and agonising decisions had to be made by the domestic authorities. It is prepared to grant them a margin of appreciation, at least in so far as the military and technical aspects of the situation are concerned, even if now, with hindsight, some of the decisions taken by the authorities may appear open to doubt.
214. In contrast, the subsequent phases of the operation may require a closer scrutiny by the Court; this is especially true in respect of such phases where no serious time constraints existed and the authorities were in control of the situation”.
The claimants in the Finogenov/Russia case argued that instead of storming the theatre, the Russian authorities should have continued to negotiate, after which people may have been released, and that the actual objective of the action was first and foremost to kill the terrorists. The ECtHR, however, ruled that the threat was very real and that it was unclear whether the terrorists would have been willing to make concessions in negotiations. The situation prior to the storming was therefore highly alarming. The Court went on to rule in para. 226:
“(…) that there existed a real, serious and immediate risk of mass human losses and that the authorities had every reason to believe that a forced intervention was the “lesser evil” in the circumstances. Therefore, the authorities’ decision to end the negotiations and storm the building did not in the circumstances run counter to Article 2 of the Convention”.
The use of the intoxicating gas, causing the death of many hostages, was not disproportionate either, according to the ECtHR, because its use contributed to the release of the other hostages and reduced the risk of the terrorists detonating the bombs they had rigged. The Court was unable to assess whether the dose of gas had been correctly calculated.87
As explained at 5.12, the ECtHR ruled that the Russian authorities had nevertheless violated their positive obligations under Article 2 ECHR in their preparation for and the planning of the operation. The Court applied closer scrutiny to this aspect of the operation:
“243. (…) the Court considers that the planning and conduct of the rescue operation, in particular the organisation of the medical aid to the victims and their evacuation, can be subjected to a more thorough scrutiny than the “political” and military aspects of the operation. The Court notes, first of all, that the rescue operation was not spontaneous: the authorities had about two days to reflect on the situation and make specific preparations. Second, in this area (evacuation and medical assistance) the authorities should have been in a position to rely on some generally prepared emergency plan, not related to this particular crisis. Third, they had some control of the situation outside the building, where most of the rescue efforts took place (contrary to the situation within the building, which was in the hands of the terrorists). Finally, the more predictable a hazard, the greater the obligation to protect against it: it is clear that the authorities in this case always acted on the assumption that the hostages might have been seriously injured (by an explosion or by the gas), and thus the large number of people in need of medical assistance did not come as a surprise. The Court considers that in such circumstances it may subject the rescue operation, in so far as it concerned the evacuation of and medical assistance to the hostages, to closer scrutiny”.
Briefly put, it follows from the foregoing that the ECtHR allows the authorities a certain margin of appreciation in the military and technical aspects when taking operational decisions, even when it can be doubted according to current insight whether the authorities took the right decision at the time. By contrast, the ECtHR can subject other phases of the relevant operation to close scrutiny. In these phases, the disappearance or absence of high time pressure under which the decisions must be taken may also play a role, also in the event that the authorities have complete control of the situation.
In connection with this, the English Supreme Court’s judgment in the case of Smith and others v. Ministry of Defence is interesting.88 This case concerns the concept of combat immunity, developed under common law, and the scope of Article 2 ECHR in armed conflict situations. Combat immunity entails immunity from liability under civil law that applies to the government and to members of the armed forces, including peacekeeping troops and the police, within the context of actual or threatened armed violence. In these cases, under common law the members of the armed forces and the government do not have an enforceable duty of care in respect of one another or anyone else to prevent damage, personal injury or loss of life.89 In the event that a defendant can demonstrate their protection under combat immunity, the court must decline jurisdiction or declare the relevant claim inadmissible. The questions at issue in the Smith case included whether this immunity also extends to the preparations for the armed conflict, such as the training and equipping of the military, during which phase there is no actual or threatened armed conflict. With a majority of four to three, the Supreme Court decided that it did not. I will quote Lord Hope, with whom the majority of the justices concurred, regarding the doctrine of combat immunity:
“94. (…) Starke J observed in the Shaw Savill case at p 354 that not every warlike operation done in time of war is an operation or an act of war. It is to operations or acts of war only that the doctrine extends, on the ground that the armed forces must be free to conduct such operations without the control or interference of the courts of law. As Dixon J said in the same case at p 361, no-one can imagine a court undertaking the trial of an issue as to whether a soldier on the field of battle or a sailor fighting on his ship might reasonably be more careful to avoid causing civil loss or damage. The principle (…) is not limited to acts or omissions in the course of an actual engagement with the enemy. It extends to all active operations against the enemy. While in the course of actually operating against the enemy, the armed forces are under no duty of care to avoid causing loss or damage to those who may be affected by what they do. But, as Dixon J also said at p 362, there is a real distinction between actual operations against the enemy and other activities of the combatant services in time of war. He referred by way of example to a warship proceeding to her anchorage or manoeuvring among other ships in a harbour. At that stage no reason was apparent for treating her officers as under no civil duty of care, remembering always that the standard of care is that which is reasonable in the circumstances.
95. (…) At the stage when men are being trained, whether pre-deployment or in theatre, or decisions are being made about the fitting of equipment to tanks or other fighting vehicles, there is time to think things through to plan and to exercise judgment. These activities are sufficiently far removed from the pressures and risks of active operations against the enemy for it to not to be unreasonable to expect a duty of care to be exercised, so long as the standard of care that is imposed has regard to the nature of these activities and to their circumstances”.
Lord Hope also devoted attention to the question of the extent to which Article 2 ECHR applies to the preparation for and execution of military operations. I quote:
“64. The extent to which the application of the substantive obligation under article 2 to military operations may be held to impossible or inappropriate will, however, vary according to the context. Military operations conducted in the face of the enemy are inherently unpredictable. There is a fundamental difference between manoeuvres conducted under controlled conditions in the training area which can be accurately planned for, and what happens when troops are deployed on active service in situations over which they do not have complete control. As Lord Rodger observed (…) the job of all members of the armed forces involves their being deployed in situations where, as they well know, opposing forces will be making a determined effort, and using all their resources, to kill and injure them. The best laid plan rarely survives initial contact with the enemy. The best intelligence cannot predict with complete accuracy how the enemy will behave, or what equipment will be needed to meet the tactics and devices that he may use to achieve his own ends. Speed may be essential if the momentum of an attack is to be maintained or to strengthen a line of defence. But lines of communication may become stretched. Situations may develop where it is simply not possible to provide troops in time with all they need to conduct operations with the minimum of casualties. Things tend to look and feel very different on the battlefield from the way they look on such charts and images as those behind the lines may have available to them. A court should be very slow indeed to question operational decisions made on the ground by commanders, whatever their rank of level of seniority”.
Lord Hope went on to note in respect of the case law of the ECtHR regarding Article 2 ECHR:
“76. The guidance which I would draw from the Court’s jurisprudence in this area is that the court must avoid imposing positive obligations on the state in connection with the planning for and conduct of military operations in situations of armed conflict which are unrealistic or disproportionate. But it must give effect to those obligations where it would be reasonable to expect the individual to be afforded the protection of the article. It will be easy to find that allegations are beyond the reach of article 2 if the decisions that were or ought to have been taken about training, procurement or the conduct of operations were at a high level of command and closely linked to the exercise of political judgment and issues of policy. So too if they relate to things done or not done when those who might be thought to be responsible for avoiding the risk of death or injury to others were actively engaged in direct contact with the enemy. But finding whether there is room for claims to be brought in the middle ground, so that the wide margin of appreciation which must be given to the authorities or to those actively engaged in armed conflict is fully recognised without depriving the article of content, is much more difficult. No hard and fast rules can be laid down. It will require the exercise of judgment. This can only be done in the light of the facts of each case”.
From the matters discussed above regarding the State’s obligations under the ECHR, I am of the opinion that it can be concluded that in the event that the State knows or ought to know that there is a real risk, as a result of the conduct of third parties, of death or inhumane treatment of certain individuals or groups of people over whom the State has legal authority or for whom the State has an obligation of care within the meaning of Article 6:162 DCC, the State has an obligation to take all reasonable measures to avoid that risk to the extent possible. That obligation also exists in situations of armed conflict in which peacekeeping troops have been deployed by the State.
In the assessment of what can reasonably be expected of peacekeeping troops in situations of armed conflict, all of the circumstances of the case are relevant. Particularly relevant is the extent to which the peacekeeping troops have control of the situation, which is understood to include the extent to which they can effectively – on site in a practical sense – influence the conduct of the third parties involved, as well as the time pressure under which the relevant decisions must be taken. Where they have the time to make preparations, the threat does not come as a surprise and they can control the situation, more may be reasonably expected of the peacekeeping troops, and the court will assess whether they did all they could reasonably be expected to do with closer scrutiny.
If action must be taken under heavy time pressure, there is little control over the situation and little if any influence can be exerted on the third parties involved, while the real risk of death or inhumane treatment is imminent, less may be reasonably expected of peacekeeping troops. This is all the more true when action must be taken within the context of actual or threatened armed violence directed against the peacekeeping troops themselves, as well. In such cases, when a choice is made between two evils, to wit a choice between two different risks – and it may be reasonably assumed based on the information available that this choice could lead to the same or a better result – and the risk taken proceeds to materialise, I am absolutely convinced that the court should not involve itself in the relevant decision-making on the ground that the other choice may have led to a better result and the peacekeeping troops therefore reasonably could have done more to alleviate the risk that was involved. Relevant is the fact that in situations of armed conflict, or in situations like the one at issue in Finogenov/Russia, it may be the case that peacekeeping troops must make a choice between actions that each entail real and even very great risks for the lives or well-being of the people placed under their protection. It is therefore possible to make a legitimate choice in such cases for a certain action despite the large probability of the risks involved actually materialising.
By assessing the conduct of peacekeeping troops in situations of armed conflict in this manner, the court does not act contrary to the judgments of the Supreme Court in the [A] and [B] cases. Therefore, in doing so the court is not acting in a manner that the Supreme Court qualifies as unacceptably reticent. After all, it is not the case that this assessment leaves virtually no margin of appreciation of the consequences of the conduct of the peacekeeping troops (see para. 3.18.3 of the Supreme Court’s judgments).
After this assessment, I return to the discussion of the principal ground for cassation.
Part I (principal ground for cassation): knowledge of the Dutchbat command
concerns the Court of Appeal’s opinion in paras. 50.1-52.7 regarding the knowledge of the Dutchbat command of the fate of the men. This ground for cassation can be divided into four sub-parts.
complains that the Court of Appeal failed to appreciate that in the application of the standard of wrongfulness at issue, the actual knowledge of the Dutchbat command at the time of the contested conduct is concerned, and not what the Dutchbat command ought to have known or understood. Furthermore, with this opinion the Court of Appeal exceeded the limits of the legal battle on appeal, because the Foundation et al. had not directed grounds for appeal against the District Court’s opinion in para. 4.180 of its final decision, which concerns whether the Dutchbat command reasonably could have decided and acted as it did, in view of what was known to the command at the time of the contested conduct.
It ensues from my general explanation above that the Court of Appeal did not start from an incorrect interpretation of the law by assessing what the Dutchbat command knew or ought to have known. Furthermore, the Court of Appeal did not exceed the limits of the legal battle on appeal, as the District Court evidently not only assessed what the Dutchbat command knew, but also what it ought to have known (see paras. 4.255, 4.257, 4.324 and 4.328 of the District Court’s final decision). The complaints fail for that reason.
is directed against para. 50.1 of the contested judgment and complains, briefly put, that the Court of Appeal wrongly ruled that the mere knowledge of Dutchbat of the real risk of inhumane treatment or execution of the men by the Bosnian Serbs means that Dutchbat’s conduct was wrongful.
The part departs from an incorrect reading of the judgment and fails for that reason. By ruling that the knowledge of a real risk that the men would be killed or inhumanely treated by the Bosnian Serbs has the same legal consequences as the knowledge of genocide, the Court of Appeal did not rule that this knowledge as such means that Dutchbat’s conduct was unlawful (also see the discussion of part 5.2 of the ground for cross-appeal in cassation). This is because the Court of Appeal assessed the lawfulness of Dutchbat’s conduct based on the question of what could be reasonably expected of Dutchbat, given that Dutchbat knew or ought to have known that a real risk existed of execution or inhumane treatment of the men by the Bosnian Serbs.
concerns the Court of Appeal’s opinion in paras. 51.3 through 51.6 that the Dutchbat command knew that the men ran a real risk of death or inhumane treatment if they were separated and deported by the Bosnian Serbs. The part complains that this opinion was insufficiently comprehensively reasoned. This part of the ground for cassation can be divided into four sub-parts.
Those sub-parts contain the following complaints:argues that in so far as the Court of Appeal based its opinion on the statements by Van Duijn, Rutten and Oosterveen cited in para. 51.4, the Court of Appeal failed to appreciate that these individuals were not part of the Dutchbat command and could not issue orders, while only Karremans and Franken were the Dutchbat commanders in the enclave. complains that if the Court of Appeal assumed that the findings of Van Duijn, Rutten and Oosterveen also reached the Dutchbat battalion command, that assumption is incomprehensible, as this does not follow from those statements, and Oosterveen also explicitly indicated that he did not pass on his observation of shooting. argues that it does follow from the statements by Franken mentioned by the Court of Appeal at b, c and f in para. 51.4 that he was concerned about the treatment of the men by the Bosnian Serbs, but it does not follow that he actually had knowledge of a real risk of inhumane treatment or execution of the men. As Franken did not draw up his list until 13 July 1995, that circumstance cannot contribute to the opinion that the aforementioned knowledge already existed on 12 July 1995. argues that alongside the aforesaid statements, the Court of Appeal merely based its opinion on the statement by Karremans in the morning of 13 July 1995 of the discovery of nine bodies and an observation of an execution. In themselves, these circumstances are insufficient to assume that the aforementioned knowledge already existed in the evening of 12 July 1995, according to this sub-part.
Contrary to the apparent point of departure of part 1.3.1, the Court of Appeal did not deem the knowledge of Van Duijn, Rutten and Oosterveen to be knowledge of the Dutchbat command. In para. 51.4 the Court of Appeal refers to the statement by Van Duijn – which does not, in itself, bear the Court of Appeal’s opinion – in which he said that the men were not deported in a truck until the end of the evening. This paragraph must be read in conjunction with what the Court of Appeal ruled in para. 51.2, where it says that Van Duijn stated before the Parliamentary Investigative Committee that he was asked whether he could still see the men, and that his answer was that he could, as they were sitting in a field of grass in front of a house. In para. 51.4d the Court of Appeal cites Rutten’s statement, where he said that the company’s commander – who himself can be deemed a leader90 – reported matters to the battalion regarding the course of events around the buses in the morning of 13 July 1995, and that Rutten personally reported the course of events around the “White House” to Karremans. In para. 51.4 the Court of Appeal cites Oosterveen’s statement, where he said that intermittent shots could be heard, from which Oosterveen apparently concluded that this was not combat fire – as that would be a rattling sound – but execution shots. These shots could be heard by all, including the Dutchbat command. In view of the foregoing, it is not incomprehensible that the Court of Appeal took the statements by Van Duijn, Rutten and Oosterveen into consideration in its assessment of the knowledge that existed among the Dutchbat leaders. To that extent, the complains in parts 1.3.1 and 1.3.2 fail.
Part 1.3.3 departs from an incorrect interpretation of the law that it must be established that Franken had actual, and not just “normative”, knowledge of a real risk of inhumane treatment or execution of the men by the Bosnian Serbs. In this regard, the Court of Appeal was certainly able to derive from the statements by Franken reflected in 51.4 at c, the excerpt from the NIOD report reflected in 51.4 at b, and the conversation with a representative of the refugees reflected in 51.4 at f, that this involved more than just the concerns that Franken allegedly had about the fate of the men, or at least that the concerns were so serious that these were concerns about the real risk of inhumane treatment or execution of the men by the Bosnian Serbs, which boils down to knowledge, normative or not, of that real risk. To that extent, therefore, part 1.3.3 fails.
The given that Franken had a list made up on 13 July 1995 does not bear the Court of Appeal’s opinion that knowledge of a real risk already existed in the evening of 12 July 1995. Furthermore, it is not incomprehensible that the Court of Appeal took that given into consideration in its assessment of the knowledge that must have existed in the evening of 12 July 1995. The complaint from part 1.3.3 also fails in this regard.
apparently assumes that the complaints from parts 1.3.1-1.3.3 would be successful. As they fail, the complaint in part 1.3.4 also fails.
embroiders upon part 1.3 and must share its fate.
Part 3 (principal ground for cassation): Dutchbat’s conduct with the “sluice” during the evacuation on 13 July 1995
is directed against paras. 61.1 through 61.8 and 65, where the Court of Appeal ruled that Dutchbat acted wrongfully by continuing to cooperate in the evacuation on 13 July 1995 by creating groups and a “sluice”, by means of which Dutchbat facilitated conduct by the Bosnian Serbs that entailed, according to what Dutchbat knew or ought to have known, the real risk that male refugees would be exposed to inhumane treatment or execution.
needs no discussion, as it embroiders upon part 1 and must share its fate.
unconditionally refrained from creating the “sluice”, as it was operating in a situation of armed conflict and had to take decisions under a large amount of pressure. In such situations, one may be forced to choose “between two evils”. Whether the choice for a certain course of action is justified in such a situation depends on the other circumstances of the case. complains that even if the Court of Appeal did not fail to appreciate this, the Court of Appeal’s opinion is incomprehensible in view of the circumstances established by the Court of Appeal in paras. 61.66 and 64.2, to wit that (i) by forming the sluice, the interests of the women, children and elderly were served; (ii) chaos would have ensued if the cooperation was withdrawn; (iii) chaotic masses of people would have ended up at the buses who would push all at once in an attempt to board the buses, which would subsequently be forced to leave overloaded; (iv) it is plausible that the Bosnian Serbs would treat the people harshly if the Dutchbat military did not stand between them and the people; (v) it is plausible that withdrawing the cooperation would not have helped and the men would have ended up in the hands of the Bosnian Serbs in that event, as well.concisely reflects the Court of Appeal’s considerations in paras. 61.1 through 61.8 and 65. This part can be divided into two sub-parts. complains that the Court of Appeal failed to appreciate that Dutchbat should have
Parts 2.2.2 and 2.2.3 can be jointly discussed. I put first and foremost that in para. 39.2 of its judgment, the Court of Appeal established that Dutchbat was operating in a situation of armed conflict and was forced to take decisions under a significant amount of pressure, and that it also follows from the Court of Appeal’s judgment that Dutchbat was generally acting under the threat of armed violence.91 In addition, it follows from paras. 37.6-38.4, where the Court of Appeal discussed the legal authority exercised by the State, that although Dutchbat exercised some degree of control inside the compound, it could not exercise it outside in the mini safe area, where the buses were located. Dutchbat could also influence the actions of the Bosnian Serbs little, if at all. This already follows from the evident opinion of the Court of Appeal that the only alternative to Dutchbat’s manner of cooperation in the evacuation was to entirely refrain from cooperating, after which the men would be forced by necessity to go to the buses themselves and would be separated by the Bosnian Serbs.92 That this manner was the very manner that Dutchbat should have chosen according to the Court of Appeal does not change the fact that Dutchbat could exert little, if any, influence on the conduct of the Bosnian Serbs. As I have explained above, all this is relevant to the assessment of Dutchbat’s conduct.
It appears from the judgment of the Court of Appeal that Dutchbat could reasonably assume that forming the sluice would help the group of Bosnian Muslims as a whole. This is because the Court of Appeal ruled that it was plausible that chaos would ensue if the cooperation was withdrawn because many thousands of people wanted to leave, and that it was plausible that the Bosnian Serbs would have treated the people harshly in that event, while the Court of Appeal established in para. 64.2 that it was not plausible that the people would have been in a better position in that event. The Court of Appeal also determined in para. 61.7 that during the cooperation, individual Dutchbat members prevented a number of older men or young boys from being separated from their families. Under those circumstances, it cannot be said that Dutchbat automatically acted unlawfully by choosing between “two evils”, namely facilitating in the separation of the men on the one hand, and a chaotic evacuation on the other hand. In my opinion, this can only be different if Dutchbat or its command knew or ought to have known with certainty that the men were being led to their death or inhumane treatment. That certainty did not exist, however. The Court of Appeal has failed to appreciate all of this. The Court of Appeal therefore departed from an incorrect interpretation of the law or insufficiently comprehensively reasoned its opinion. It is therefore my opinion that the complaints in parts 2.2.1 and 2.2.2 are successful.
In view of the foregoing, the complaint in, which embroiders upon parts 2.2.1 and 2.2.2, concerning the paras. 64.1 through 65, paras. 73.1 through 73.3 and the operative part of the Court of Appeal’s judgment, is successful to the extent that it concerns the circumstance of the creation of a sluice by Dutchbat.
Part 3 (principal ground for cassation): Dutchbat’s conduct in respect of the men who were inside the compound on 13 July 1995
concerns the Court of Appeal’s opinion that Dutchbat acted wrongfully by not explicitly giving the men who were inside the compound on 13 July 1995 the choice, also referring to the risks they would be exposed to if they left with their families, of staying behind while the women, children and elderly were being evacuated from the compound. This part can be divided into two sub-parts.
contains a complaint that embroiders upon part 1, and fails for that reason.
complains that, already in light of the chaotic circumstances, the Court of Appeal’s assumption that there was a real possibility for Dutchbat to use a megaphone in the factory hall to warn the approx. 320 men amongst the 5,000 refugees is incomprehensible. That assumption is also incomprehensible in view of the State’s position that the refugees would not have accepted a separation of the men inside the compound. This means that the Court of Appeal’s opinion in para. 63.2, that Dutchbat could have created a “void” between the refugees by warning the men inside the compound, and that the men would then voluntarily stay behind, is also incomprehensible. complains that it already ensues from what was put forward in part 3.2 that Dutchbat was reasonably entitled to decide not to offer the men the choice of staying behind in the compound.
Contrary to what this part argues, the Court of Appeal’s opinion is not based on the assumption that Dutchbat should have used a megaphone in the factory hall. This is because it ensues from para. 63.2 that the Court of Appeal assumed that Dutchbat had sufficient time to talk with the men and explain where necessary, by means of the interpreters that were present or through the committee formed on site that represented the refugees. Nor does the Court of Appeal’s opinion infer that Dutchbat subsequently should have used violence or should have forced the men to stay if they did not do so voluntarily. After all, in para. 63.2 the Court of Appeal holds that it is disregarding these arguments put forward by the State, as in any event Dutchbat could have explained the risks and given the men the choice of staying behind in the compound. That opinion is not incomprehensible, and part 3.2 fails for that reason. Part 3.3, which entirely embroiders upon part 3.2, therefore also fails.
concerns the Court of Appeal’s opinion in para. 63.4 of its judgment that the State had insufficiently substantiated that the scarcity of water, food, medicine and sanitary facilities was so urgent that the State could already reasonably decide to also cooperate in the evacuation of the group of men inside the compound as soon as on 13 July 1995. This part complains that the Court of Appeal has failed to appreciate that the relevant obligation to furnish facts is borne not by the State, but by the Foundation et al. as the claimants.
The Court of Appeal did not fail to appreciate that the obligation to furnish facts that led to the opinion that Dutchbat acted wrongfully is borne by the Foundation et al., but evidently ruled that in light of the given that the male refugees comprised only a small part of the refugees that were inside the compound, the State had insufficiently disputed with substantiation the arguments of the Foundation et al. entailing that it was possible to keep the approx. 320 male refugees inside the compound longer. The complaint therefore fails.
complains that the Court of Appeal in any event insufficiently reasoned its opinion regarding the keeping of the male refugees behind in the compound, in light of the State’s refutation of the assertions made by the Foundation et al. regarding that possibility. Within that context, the State argued that (i) the circumstances inside the compound were already deplorable as from 11 July 1995; (ii) on 13 July 1995, there was still no insight into when Dutchbat would be able to leave; (iii) on 13 July 1995 the UN convoys with food and other stocks had been detained by the Bosnian Serbs; (iv) the size of the group of men involved was so large that Dutchbat was reasonably able to decide that the stocks available at that time were woefully insufficient for keeping the group of men behind. According to the judgment, the Court of Appeal’s endorsed most of these arguments, as it held that (i) the circumstances were horrible on 11 July 1995 and the refugees had no access to food and water; (ii) these circumstances visibly deteriorated on 12 and 13 July 1995; (iii) people had been buried inside the compound who had died of exhaustion; (iv) the weather did not improve on 12 and 13 July 1995; (v) the shortages could not be supplemented; and (vi) under such circumstances, people will only survive for a few days.
The complaint of part 3.4.2 fails. The opinion of the Court of Appeal that it was possible to still keep the men inside the compound for some time, even under the deplorable and chaotic circumstances inside the compound, is not incomprehensible in itself, as the Court of Appeal held in para. 63.4 that the male refugees comprised only a fraction of the total number of refugees, that it was established that there was still a bit of drinking water, and that the documents say nothing about any outbreak of disease.
complains that in view of these circumstances, the Court of Appeal could not arrive at the conclusion, without additional reasoning, that Dutchbat acted wrongfully by not offering the male refugees the option of staying behind in the compound.
complains that the Court of Appeal failed to appreciate that already in the early and late afternoon of 13 July 1995, under chaotic circumstances, Dutchbat had to take a decision about the evacuation of everyone who was inside the compound, including the men, and that it ensues from the arguments put forward in parts 3.2-3.4 that Dutchbat reasonably decided not to allow the men to stay behind in the compound. This cannot change the fact that Dutchbat could have “delayed” this decision until a later point in time.
I will discuss parts 3.4.3 and 3.5 jointly. In the assessment of whether Dutchbat acted wrongfully, it is relevant that Dutchbat was operating in a situation of armed conflict and was forced to take decisions under a significant amount of pressure93, and that it also follows from the Court of Appeal’s judgment that Dutchbat was generally acting under the threat of armed violence. Unlike in the mini safe area, Dutchbat did have a certain degree of control of the events inside the compound. Furthermore, Dutchbat could exercise some, limited, influence on the conduct of the Bosnian Serbs, in the sense that Dutchbat could have kept the male refugees inside the compound for some time and in any event could have postponed the time at which the men fell into the hands of the Bosnian Serbs. In view of the latter, it is my opinion that the Court of Appeal was right to consider whether Dutchbat – with the knowledge it had at that time – could have reasonably decided to take a course of action that, being assessed with hindsight, might have led to more favourable results for the male population than the course of action it actually did take, the results of which are known. This is not what the complaint is about, either. In the assessment of what Dutchbat could be reasonably expected to do, serious weight is nevertheless carried by the fact that Dutchbat was acting in a situation of armed conflict, under significant pressure and under the threat of armed violence, with only a very limited possibility of influencing the Bosnian Serbs.
The Court of Appeal based its opinion on the wrongfulness of the Dutchbat’s conduct on its opinion that Dutchbat could have kept the male refugees inside the compound for a few days, or even a few hours, more. However, it ensues from the Court of Appeal’s opinion in paras. 66.2-69.1 that even if that course of action had been followed, the male refugees ran a very high risk of being killed or treated inhumanely. In addition, following that course of action would entail the men staying behind in a deplorable situation, which was visibly deteriorating, while, as argued by the State, there was no insight on 13 July 1995 as to when Dutchbat itself would be able to leave and the stock of supplies was being detained. In that light, the Court of Appeal has insufficiently reasoned why Dutchbat could not reasonably decide on 13 July 1995, weighing the real risks that were involved in both courses of action and given the situation Dutchbat was in at that time, to also evacuate the men. It is therefore my opinion that the complaints in parts 3.4.3 and 3.5 are successful.
contains an embroidering complaint. It argues that the success of one or more of the complaints from part 3 must also mean that the embroidering opinion of the Court of Appeal in paras. 66.1 through 69.1 cannot be upheld. In those paragraphs, the Court of Appeal assessed the question of whether and to what extent there is a causal connection between the conduct of which Dutchbat is accused and the fate of the men inside the compound. This part also argues that the Court of Appeal’s embroidering opinions in respect of the men inside the compound and the damage suffered by their surviving dependants (paras. 71.2, 73.1 through 73.3, 75 and the operative part) cannot be upheld.
In light of the success of the complaints from part 3.4.3 and from part 3.5, the embroidering complaint from part 3.6 is also successful.
Part 4 (principal ground for cassation): causal link between the conduct of which Dutchbat is accused and the fate of the men inside the compound
complains about the causal connection between the conduct of which Dutchbat is accused and the fate of the men inside the compound. After an introduction, the part can be divided into five sub-parts.
As the complains from parts 3.4.3, 3.5 and 3.6 are successful, a discussion of part 4 is unnecessary as matters stand.
Part 5 (principal ground for cassation): claim for damages, to be determined in follow-up proceedings
concerns the order for the State to compensate the damage suffered by the wrongful conduct of Dutchbat, to be determined in follow-up proceedings and settled in accordance with the law. This part complains that the opinion of the Court of Appeal is incomprehensible in so far as the Court of Appeal intended to pronounce this order in respect of the ten individual claimants, and incorrect in so far as the Court of Appeal intended to pronounce the order in respect of the Foundation.
The claimants’ requests were denied because their men were not evacuated from inside the compound on 13 July 1995. The order for damages, to be determined in follow-up proceedings, therefore may only be pronounced in respect of the Foundation. However, pursuant to Article 3:305a(3) DCC, the Foundation does not have the power to institute a legal claim seeking a monetary amount in damages. The complaint in part 5 is therefore successful.
The final conclusion is that the principal ground for cassation is successful.
6. Conclusion in the principal ground for cassation and the ground for cross-appeal in cassation
The conclusion is that the case in the principal appeal should be quashed, with referral, and that the cross-appeal in cassation should be rejected.
The Procurator-General with the
Supreme Court of the Netherlands
1 Supreme Court, 13 April 2012, ECLI:NL:HR:2012:BW1999, NJ 2014/262, annotated by N.J. Schrijver to NJ 2014/263.
2 ECtHR, 11 June 2013, no. 65542/12, NJ 2014/263, annotated by N.J. Schrijver.
3 See ECLI:NL:GHDHA:2017:1761, NJF 2017/466; ECLI:NL:GHDHA:2017:3376 for an English translation of the judgment.
4 See Supreme Court, 6 September 2013, ECLI:NL:HR:2013:BZ9228, NJ 2015/376, annotated by N.J. Schrijver ([A]) and Supreme Court, 6 September 2013, ECLI:NL:HR:2013:BZ9225, RvdW 2013/1037 ([B]).
5 See paragraph 10 of the Foundation et al.’s written explanation, p. 35.
6 The ILC adopted the DARS during its 53rd session in 2001. DARS, with accompanying commentary, was published in: Yearbook of the International Law Commission 2001, vol. II, Part Two.
7 The ILC adopted the DARIO during its 63rd session in 2011. DARIO, with accompanying commentary, was published in: Yearbook of the International Law Commission 2011, vol. II, Part Two.
8 Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), Merits, Judgment, ICJ Reports 1986, p. 14.
9 See also Jane M. Rooney, The Relationship between Jurisdiction and Attribution after Jaloud v. Netherlands, NILR 2015/3, pp. 422-23.
10 Case Concerning the Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro), Judgment, ICJ Reports 2007, p. 43.
11 R. Kolb, The International Law of State Responsibility, 2017, p. 81.
12 Prosecutor v Tadić, IT-94-1-A, Judgment, 15 July 1999.
13 Antonio Cassese, The Nicaragua and Tadić Tests Revisited in Light of the ICJ Judgment on Genocide in Bosnia, EJIL 2007/4, p. 654. See also Jure Vidmar, Some Observations on wrongfulness, Responsibility and Defences in International Law, NILR 2016/3, p. 344, et seq. and Kolb, op. cit., p. 105, et seq., with references to the literature. It is worth noting that even if a State is not responsible for the harm caused by the conduct of persons in other States, it may nevertheless be held responsible for its own conduct in that regard, as the US was held responsible in the Nicaragua case for its own violations of international law.
14 Tallinn Manual 2.0 on the International Law Applicable to Cyber Operations. Prepared by the International Groups of Experts at the Invitation of the NATO Cooperative Cyber Defence Centre of Excellence, Cambridge: 2017, p. 96.
15 See also Tom Dannenbaum, Translating the Standard of Effective Control into a System of Effective Accountability: How Liability Should Be Apportioned for Violations of Human Rights by Member State Troop Contingents Serving as United Nations Peacekeepers, Harv. ILJ 2010/1, p. 157: ‘(…) effective control is held by the entity that is best positioned to act effectively and within the law to prevent the abuse in question’.
16 Tom Dannenbaum, Killings at Srebrenica, Effective Control, and the Power to Prevent wrongful Conduct, ICLQ 2012, p. 723.
17 In his 2010 publication, Dannenbaum (op. cit., p. 156) refers to the issue of attributing the conduct of private persons as being ‘slightly different’ to the issue of attributing conduct in the context of peacekeeping missions.
18 In that vein, see also Natalia Perova, Disentangling ‘Effective Control’ Test for the Purpose of Attribution of the Conduct of UN Peacekeepers to the States and the United States, Nordic Journal of International Law 86 (2017), p. 53, et seq.
19 With regard to the term ‘command and control’ and the consequences of the transfer of same, see, inter alia, B. van Lent, “Command and Control” in VN-vredesoperaties, Traktatie 1995, pp. 19-22; Terry D. Gill, Legal Aspects of the Transfer of Authority in UN Peace Operations, NYIL 2011, p. 45, et seq.; Dannenbaum (2010), op. cit., pp. 142-51; Ulf Häuβler, Human Rights Accountability of International Organisations in the Lead of International Peace Missions, in: J. Wouters et al. (eds.), Accountability for Human Rights Violations by International Organisations, 2010, p. 233, et seq.
20 See paragraph 9 of the Commentary to Article 7 DARIO, quoting the Secretary-General of the UN: ‘The international responsibility of the United Nations for combat-related activities of United Nations forces is premised on the assumption that the operation in question is under the exclusive command and control of the United Nations. (…) In joint operations, international responsibility for the conduct of the troops lies where operational command and control is vested according to the arrangements establishing the modalities of cooperation between the State or States providing the troops and the United Nations’.
21 See paragraph 9.3 of the Foundation et al.’s written explanation.
22 See also Ömer Frank Direk, Responsibility in Peace Support Operations: Revisiting the Proper Test for Attributing Conduct and the Meaning of the ‘Effective Control’ Standard, NILR 2014, p. 19, which indicates that the rules on command and control have a fundamental impact on the question of who has effective control. In a comparable sense, see Bérénice Boutin, Responsibility of the Netherlands for the Acts of Dutchbat in [B] and [A], Leiden J. of Int. Law 2012/2, p. 528; Perova, op. cit., p. 45. Dannenbaum (2010; op. cit., p. 157) also asserts that ‘those command and control powers most “relevant” to a given conduct are those most likely to be useful in preventing that conduct from occurring’.
23 See also Ray Murphy & Siobhán Wills, United Nations Peacekeeping Operations, in: André Nollkaemper & Ilias Plakokefalos, The Practice of Shared Responsibility in International Law, Cambridge: 2017, p. 599: ‘Establishing effective control requires a high standard of proof’.
24 Cf. Direk, op cit., p. 16-17, who asserts that the ‘preventative rationale’ of the standard proposed by Dannenbaum may not be interpreted broadly because otherwise virtually all acts of UN peacekeeping troops would be attributable to the sending State. According to Direk, therefore, dual attribution must be based on stricter grounds (p.16): ‘In particular, the issue of dual attribution should be conducted within stricter limits, even though it may be tempting, in seeking to provide as much protection as possible for victims of wrongful acts, to opt for a wide interpretation that attributes responsibility for the impugned conduct to as many actors as possible’.
25 Also according to André Nollkaemper, Dual Attribution. Liability of the Netherlands for Conduct of Dutchbat in Srebrenica, Journal of International Criminal Justice 2011/9, p. 1148, et seq.
26 The Hague Court of Appeal 5 July 2011, ECLI:NL:GHSGR:2011:BR0132 and ECLI:NL:GHSGR:2011:BR0133.
27 See paragraph 5.2.3, et seq., of the Foundation et al.’s written explanation.
28 Dannenbaum (2010), op. cit., p. 160.
29 Cedric Ryngaert, Peacekeepers Facilitating Human Rights Violations: The Liability of the Dutch State in the Mothers of Srebrenica Cases, NILR 2017, p. 455; Paolo Palchetti, The allocation of responsibility for internationally wrongful acts committed in the course of multinational operations’, International Review of the Red Cross 2013, pp. 734-735. See also paragraphs 3.3.6-3.3.9 of the State’s written explanation.
30 See DARIO, Commentary, Yearbook of the ILC 2011, vol. II, Part Two, 61: ‘Practice of international organizations confirms that ultra vires conduct of an organ or agent is attributable to the organization when that conduct is linked with the organ’s or agent’s official functions’.
31 Cf. Bérénice Boutin, The Role of Control in Allocating International Responsibility in Collaborative Military Operations, (dissertation. UvA, 2015), pp. 133-134; Perova, op. cit., p. 45, et seq.
32 DCBC stands for Defence Crisis Management Centre, see also para. 2.22 of my opinion.
33 Part 3 of the ground for cross-appeal in cassation, which is discussed below, contains still more separate complaints about this issue.
34 Cf. paras. 16.10.6, 16.14.2 and 16.14.3 of the reply, which is cited in para. 77 of the Foundation et al.’s statement of appeal in case no. 200.160.317/01.
35 Cf. paras. 77 and 300 of the Foundation et al.’s reply, as well as paras. 16.12.2 and 16.20.10 of the reply. The other sources referred to in this part (in footnote 58 – specifically, paras. 60, 63, 166, 168 and 364 of the summons, paras. 26 and 27 of the Foundation et al.’s written arguments in the first instance, and para. 68 of the Foundation et al.’s statement of appeal – contain no assertion regarding the instructions that were specifically given to Nicolai.
36 Even the sources referred to in part 1.10.1 in footnote 61 contain no such indicators.
37 See Yearbook of the ILC, 2011, vol. II, Part Two, p. 98.
38 Act of 2 July 1964, Official Gazette 964, 243, withdrawn by Act of 19 June 2003, Official Gazette 2003, 270 (International Crimes Act (Wet internationale misdrijven)).
39 Vienna Convention on the Law of Treaties, 23 May 1969, Bulletin of Treaties 1985, 79.
40 Supreme Court 1 April 2011, ECLI:NL:HR:2011:BP3044, NJ 2011/354, annotated by M.R. Mok; Supreme Court 10 October 2014, ECLI:NL:HR:2015:2928, NJ 2015/12, annotated by E.A. Alkema.
41 See Jan Klabbers, Verdragenrecht, in: Nathalie Horbach, René Lefeber, Olivier Ribbelink (eds.), Handboek Internationaal Recht, 2007, p. 127. Also see Supreme Court 24 November 2017, ECLI:NL:HR:2017:2992, RvdW 2017/1257, para. 3.4.2 and my opinion for that judgment (ECLI:NL:PHR:2017:929) at no. 2.10; Supreme Court 15 June 2018, ECLI:NL:HR:2018:918, RvdW 2018/727, para. 3.5.
Convention on the Prevention and
Punishment of the Crime of G, adopted in Paris on 9 December 1948, Bulletin of Treaties 1960, 32 (authentic English and French texts as well as the Dutch translation; the authentic Chinese, Spanish and Russian texts were not included in the Bulletin of Treaties).
43 Also see William A. Schabas, Genocide in International Law, 2009, p. 405: “The Genocide Convention provisions cannot easily be applied within domestic law without some additional legislation and are therefore, in a general sense, not self-executing”. In a comparable sense, see: Christian J. Tams, Lars Berster & Björn Schiffbauer, Convention on the Prevention and Punishment of the Crime of Genocide: A Commentary, 2014, p. 220: ‘(…) the Genocide Convention is not self-executing’, and in footnote 17, ibid: ‘Yet some parts can indeed be regarded as self-executing, foremost the duty to annul immunities for pertinent persons as reflected by Article IV (…)’.
44 See the Bosnian Genocide judgment cited above, paras. 425 and 427.
45 In the same sense, also see for example Schabas, a.w., p. 533; Eyal Mayroz, The legal duty to ‘prevent’: after the onset of ‘genocide’, Journal of Genocide Research 2012/1, pp. 81, 89-90; Orna Ben-Naftali, The Obligations to Prevent and to Punish Genocide, in: Paola Gaeta (ed.), The UN Genocide Convention: A Commentary, 2009, p. 33, where the obligation to prevent genocide is described as “a morally pregnant but a normatively empty concept”.
46 See parts 16.2-16.3.5 of the written explanation of the Foundation et al. and parts 4.2.2-4.2.5 of the State’s written explanation.
47 To be distinguished from the attribution of external knowledge, such as attributing the knowledge of an expert to its principal.
48 See Supreme Court 6 April 1979, ECLI:NL:HR:1979:AH8595, NJ 1980/34, annotated by C.J.H. Brunner (Kleuterschool Babbel); Supreme Court 11 May 1990, ECLI:NL:HR:1990:AC2014, NJ 1990/544; Supreme Court 11 March 2005, ECLI:NL:HR:2005:AR7344, NJ 2005/576; Supreme Court 11 November 2005, ECLI:NL:HR:2005:AT6018, NJ 2007/231, annotated by J.B.M. Vranken.
49 In this regard, also see Asser/Maeijer&Kroeze 2-I* 2015/93; Asser/Tjong Tjin Tai 7-IV 2018/219; B.M. Katan, Toerekening van kennis aan rechtspersonen, dissertation RU Nijmegen, 2017, p. 164.
50 See pp. 3 and 25 of the Report of the Parliamentary Inquiry Srebrenica (Examination of witnesses). A platoon is a part of a company; also see .
51 See Asser Procesrecht/Asser 2017/303: “This concerns the determination of a fact, which is to be done based on a full evaluation of the available evidence. The ‘prima facie’ nature of the determination does not concern the assessment standard, but rather pertains to the status of the determination itself. This leaves room, for example as a result of the nature or the scope of the available evidence, for the possibility that this may prove to be different if there is proof to the contrary.”
52 Cf. Lindenbergh, T&C BW, Article 6:95, note 2.
53 Cf. T.F.E. Tjong Tjin Tai, De schadestaatprocedure, 2012, no. 423; Supreme Court 12 January 2001, ECLI:NL:HR:2001:AA9431, NJ 2001/453
54 Supreme Court 27 October 2017, ECLI:NL:HR:2017:2774, NJ 2018/152, with commentary from S.D. Lindenbergh.
55 See also the annex to the rejoinder in cassation.
56 See also paragraph 18.6.4 of the Foundation et al.’s written explanation.
57 See, in this respect, part 5 of the ground for cassation in the principal appeal.
58 Parliamentary Papers II 1991/1992, 22 486, no. 3, p. 30; Parliamentary Papers II 2009/2009, 31 762, no. 1, p 2-3; Parliamentary Papers II 2016/2017, 34 608, no. 3, p. 2-3 and 6.
59 Supreme Court 13 October 2006, ECLI:NL:HR:2006:AW2082, NJ 2008/529, with commentary from C.C. van Dam. Also see Asser Procesrecht/Giesen I 2015/207; N.E. Groeneveld-Tijssens, De verklaring voor recht, 2015, nos. 72 (footnote 2) and 73.
60 See, for example, in the context of a collective action on liability for a misleading prospectus: Supreme Court 27 November 2009, ECLI:NL:HR:2009:BH2162, NJ 2014/201, with commentary from C.E. du Perron (World Online).
61 See K. Rutten, J.R. Hurenkamp, Kroniek collectieve acties en schikkingen 2017, in: Y. Borrius et al. (eds.), Geschriften vanwege de Vereniging Corporate Litigation, 2017-2018, no. I.3.2.5.
62 Also see A.C.W. Pijls, Het causaliteitsvereiste bij prospectusaansprakelijkheid, Ondernemingsrecht 2009/41, no. 4.1, where the author points out that ‘(t)he causal connection only obtains form and substance when it is linked to a certain ground in a convention’.
63 Parliamentary Papers II 2016-2017, 34 608, no. 3, p. 2-3; see also Parliamentary Papers II 2008-2009, 31 762, no. 1, p. 2-3 (in the context of the evaluation of the Class Actions Settlement of Large-scale Losses or Damage Act).
64 See also: Parliamentary Papers II 1991/1992, 22 486, no. 3, p. 30; T.M.C. Arons et al., Inleiding, in T.M.C. Arons (eds.), Collectief schadeverhaal, 2018, p. 7; J. de Bie Leuveling Tjeenk & J.B. van der Velden, Collectief schadeverhaal vanuit het perspectief van de aangesproken partij, in: T.M.C. T.M.C. Arons (ed.), Collectief schadeverhaal, 2018, p. 94; R.M. Hermans, De oorzaken van het niet tot stand komen van collectieve schikkingen in massaschadezaken, in: M. Holtzer et al. (ed.), Geschriften vanwege de vereniging Corporate Litigation 2014-2015, p. 375.
65 See also Etienne Ruvebana, Prevention of Genocide under International Law, 2014, p. 204.
66 Cf. Article 3 of the Draft Articles on Prevention of Transboundary Harm from Hazardous Activities: ‘The State of origin shall take all appropriate measures to prevent significant transboundary harm or at any event to minimize the risk thereof’. This article is explained as follows: ‘(7) The obligation of the State of origin to take preventive or minimization measures is one of due diligence. It is the conduct of the State of origin that will determine whether the State has complied with its obligation under the present articles. The duty of due diligence involved, however, is not intended to guarantee that significant harm be totally prevented, if it is not possible to do so. In that eventuality, the State of origin is required, as noted above, to exert its best possible efforts to minimize the risk. In this sense, it does not guarantee that the harm would not occur’. Report of the International Law Commission on the work of its fifty-third session, 23 April - 1 June and 2 July - 10 August 2001, Official Records of the General Assembly, Fifty-sixth session, Supplement No.10, Document A/56/10, p. 154.
67 See also André Nollkaemper, ‘Failures to protect’ in international law, in: M. Weller (Ed.), The Oxford handbook of the use of force in international law, 2015, p. 437-461.
68 Nienke van der Have, The Prevention of Gross Human Rights Violations under International Human Rights Law, 2018, p. 182-183.
69 See also Ruvebana, op. cit., p. 196 et seq.
70 See also Tams/Berster/Schiffbauer, op. cit., p. 49; Ruvebana, op. cit., p. 107-112.
71 In the context of extradition in criminal matters, ECtHR 7 July 1989, No 14038/88, NJ 1990/158, with commentary from E.A. Alkema (Soering/United Kingdom) and ECtHR 26 June 1992, No 12747/87, NJ 1995/258, with commentary from G. Knigge (Drozd and Janousek/France and Spain). In the context of the deportation of asylum seekers: ECtHR 21 January 2011, No 30696/09, EHRC 2011/42 with commentary from A. Woltjer (M.S.S./Belgium and Greece) and ECtHR 7 March 2000, No 43844/98, JV 2000/103 (T.I./United Kingdom).
72 ECtHR 2 March 2010, No 61498/08, EHRC 2009/106, with commentary from M. den Heijer (Al-Saadoon and Mufdhi/UK).
73 ECtHR, 7 July 1989, No 14038/88, NJ 1990/158, with commentary from E.A. Alkema (Soering/United Kingdom).
74 ECtHR 28 October 1998, No 23452/94, JB 1999/25, with commentary from A.W. Heringa (Osman/United Kingdom). For the different applications of the standard formulated in Osman/United Kingdom, see: Franz Christian Ebert & Romina I. Sijniensky, Preventing Violations of the Right to Life in the European and the Inter-American Human Rights Systems: From the Osman Test to a Coherent Doctrine on Risk Prevention?, Human Rights Law Review 2015, p. 343-368.
75 ECtHR 28 March 2000, No 22492/93, EHRC 2000/37 with commentary from A.W. Heringa (Kilic/Turkey).
76 ECtHR 9 January 1995, No 22535/93 (Mahmut Kaya/Turkey).
77 Mahmut Kaya/Turkey, par. 87.
78 Kilic/Turkey, par. 68. Cf. ECtHR 28 September 2006, no. 44587/98 (Isaac/Turkey), in which the ECtHR ruled that Turkey had violated the positive obligation under Article 2 of the ECHR by not preventing Isaac from being beaten to death by Turkish demonstrators during a violent demonstration in the buffer zone between Northern and Southern Cyprus. It was established in the case that Turkish and Turkish-Cypriot police officers had taken part in the violence.
79 ECHR 13 April 2017, No 26562/07, NJ 2018/236, with commentary from M. Kuijer, par. 482.
80 ECtHR 20 December 2011, nos. 18299/03 and 27311/03, EHRC 2012/58.
81 In paragraph 265 the Court also refers to its decision of of 28 July 1998, no. 66/1997/850/1057, Ergi/Turkey, para. 81.
82 See, for example, Noam Lubell, Challenges in applying human rights law to armed conflict, International Review of the Red Cross 2005/87, p. 738.
83 Advisory Opinion, Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory, ICJ Reports 2004, p. 46.
84 ECtHR 18 September 2009, no. 16064 et seq./90, para. 185.
85 ECtHR 16 September 2014, no 29750/09.
86 However, cf. para. 164 of the ECtHR’s judgment of 7 July 2011, no. 55721/07, in the case Al-Skeini/United Kingdom, where the Court ruled that the procedural obligation from Article 2 ECHR (to conduct an effective and official investigation) ‘continues to apply in difficult security conditions, including in a context of armed conflict (…). It is clear that where the death to be investigated under Article 2 occurs in circumstances of generalised violence, armed conflict or insurgency, obstacles may be placed in the way of investigators and, as the United Nations Special Rapporteur has also observed (…), concrete constraints may compel the use of less effective measures of investigation or may cause an investigation to be delayed (…). Nonetheless, the obligation under Article 2 to safeguard life entails that, even in difficult security conditions, all reasonable steps must be taken to ensure that an effective, independent investigation is conducted into alleged breaches of the right to life (…)’.
87 Finogenov/Russia, paras. 234-235.
88  UKSC 41;  3 W.L.R. 69.
89 Mulcahy v Ministry of Defence  EWCA Civ 1323; Agnieszka Wieczorek, The Doctrine of Combat Immunity, Duty of Care and Legal Certainty in the Aftermath of the UKSC Decision in Smith v Ministry of Defence, University of Liverpool Law Review 2016/1, p. 63 et seq.; Thomas Tugendhat & Laura Croft, The Fog of Law. An introduction to the legal erosion of British fighting power’, Policy Exchange 2016, p. 31, consulted via policyexchange.org.uk.
90 A battalion is normally comprised of three companies; see .
91 For example, see paras. 2.39, 2.40, 17.2-17.4, 38.4 of the judgment.
92 Cf. paras. 61.4, 61.5, 64.2 and 65 of the judgment.
93 Also see paras. 5.24-5.27 and 5.45 of my opinion.