Between early March 1994 and late July 1995, the Dutch armed troops (Dutchbat) operated in the Srebrenica region of Bosnia and Herzegovina as part of the United Nations Protection Force (UNPROFOR).2 They were on a mission to provide humanitarian assistance and end the conflict. The Dutchbat headquarters had been set up in Potočari (the ‘compound’), which was situated at approximately five kilometres from the city of Srebrenica.3
Ensuring the safety in the Srebrenica region was never easy, despite calls by the Security Council and the presence of the Dutchbat. The situation had worsened particularly since 6 July 1995 when Bosnian Serbs, under the command of colonel-general Ratko Mladić, launched an attack on the Srebrenica safe area. Lacking any military resistance by Dutchbat, on 11 July 1995 Bosnian Serbs captured the city of Srebrenica.4 Earlier that afternoon, thousands of Bosniacs left the city and headed towards the compound in Potočari, where Dutchbat set up a mini safe area consisting of: (i) the compound; and (ii) the area in its vicinity (‘mini safe area’).5
These Bosniacs, depending on where they fled, could be categorised into three groups. The first group of about 15,000 Bosniac males fled to the nearby woods (in other words, they escaped on their own).6 About 6,000 of these males fell into Bosnian Serb hands,7 and were later found dead or are missing to this date.8 The others survived. The second group consisted of about 5,000 people who were housed inside the compound,9 of which, approximately 350 were males.10 The third group of thousands of Bosniacs were in the part of the mini safe area outside the compound, of which 3,000 were males.11
On 11 and 12 July 1995, Mladić asked the Dutchbat to surrender Bosniac males, and said that all males between the ages of 16 and 60 would be screened for war crimes.12 Dutchbat agreed to this demand.
On 12 and 13 July 1995, on the orders of the Bosnian Serbs buses arrived at the mini safe area to remove Bosniacs. As people were making their way to the buses, Bosnian Serbs pulled males out of the rows of people.13 Dutchbat was involved in separating males outside the compound, and, on 13 July 1995, also removed the approximately 350 males from inside the compound.14 These males were later found dead or are missing to this date.15 Bosnian Serbs killed in and around Srebrenica between 7,000 and 8,000 Bosniac males of all ages and raped other hundreds in a matter of days.16 This constituted genocide.17
The surviving relatives of the Srebrenica genocide victims have litigated against the Dutch State in three cases regarding the failure of the Netherlands to protect the right to life and the prohibition of torture of the Bosniac male victims.18 One of them includes the landmark case of the Mothers of Srebrenica v The Netherlands. The case was brought in the interests of more than 6,000 relatives of the Bosniac male victims.19
In 2019, the Dutch Supreme Court established that the Netherlands had acted unlawfully only in relation to one single act and concerning a small number of victims. This single unlawful act relates to the failure of Dutchbat to allow the approximately 350 Bosniac males to remain inside the compound on 13 July 1995 given the danger to their lives from Bosnian Serbs.20 More specifically, it found that ‘the liability of the State is limited to 10% of the damage suffered by the surviving relatives of these male refugees’.21 The ruling suggests there was a slim 10% chance that the approximately 350 Bosniac males would have survived had Dutchbat let them stay inside the compound. This means that, even if they had remained inside the compound, they would have faced a 90% chance of being killed by Bosnian Serbs.
The Dutch courts found no breach in relation to other alleged failures on part of Dutchbat, such as the failure to prevent the fall of Srebrenica, to protect thousands of other Bosniac males who had fled to the woods or had been prevented from entering inside the compound, the failure to report war crimes to the UN, and the act of separating Bosniac males inside and outside the compound and handing them over to the Bosnian Serbs. In relation to some of these claimed failures that fell under the jurisdiction of the Netherlands, the rationale of the Supreme Court was that, even if Dutchbat had acted differently, these Bosniac males would have been killed anyway.
I will argue that the approach taken by the Dutch Supreme Court, focusing alone on the fate of the Bosniac males – ‘they would not have survived anyway’ – is not compatible with standards of the duty to protect under the right to life and the prohibition of torture as protected by Articles 2 and 3 ECHR. It also suggests that the Dutch Supreme Court did not carefully examine the decision-making, planning and operations as required by the case-law of the European Court of Human Rights (ECtHR) regarding the positive obligation to protect in challenging security situations.
These arguments are presented in the following order. First, I will show that Articles 2 and 3 of the ECHR were applicable in the present case. Next, I will outline two ECHR requirements under the positive obligation to protect, namely (i) the duty of care or means and (ii) the decision-making, planning and operations, and argue that these standards have not been diligently applied and observed in Mothers of Srebrenica. The paper concludes by reflecting on the legacy of the Mothers of Srebrenica case regarding the standards of protection under Articles 2 and 3 of the ECHR in challenging security situations.22
The plaintiffs in Mothers of Srebrenica raised, among others, issues regarding the Netherlands’ failure to protect under Articles 2 and 3 of the ECHR.23 To activate the State’s positive obligation to protect under these Convention provisions two aspects should be established. First, whether the Netherlands had jurisdiction within the meaning of Article 1 of the ECHR, and second, whether it knew or ought to have known about the risk to life and ill-treatment of Bosniac males within the meaning of Articles 2 and 3.
With regard to jurisdiction, the domestic courts established that, after the fall of Srebrenica – from about 11 p.m. on 11 July 1995 – the Netherlands exercised effective control over Dutchbat.24 The Netherlands also exercised jurisdiction inside the compound within the meaning of Article 1 of the Convention.25 However, the Dutch courts also assessed the acts of Dutchbat outside the compound, implying that jurisdiction was not questioned with regard to Bosniacs who had remained there.26
Regarding the activation of Articles 2 and 3 of the ECHR, the ECtHR has established that ‘not every claimed risk to life can entail for the authorities a Convention requirement to take operational measures to prevent that risk from materialising’.27 However, the positive obligation of a State arises
‘where it has been established 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 avoid that risk’.28
In Osman v the United Kingdom, the Strasbourg Court established that Article 2 of the Convention is activated from the moment a State knew or ought to have known of a risk to life. In Tagayeva and others v Russia, the Strasbourg Court has clarified that such a positive obligation to protect may apply not only to one or more individuals but also to society as a whole.29 This finding indicates that Articles 2 and 3 of the ECHR were engaged in Mothers of Srebrenica if it can be established that Dutchbat knew or ought to have known of the risk to life of Bosniac males.
The Dutch classification of important documents related to Srebrenica as confidential has led to scarce and at times contradictory information about the knowledge that the Dutchbat had regarding the risk to life of Bosnian males. For example, in other proceedings regarding Srebrenica, Dutchbat had stated ‘that they had believed the [Bosnian Serbs] would treat the [Bosniac males] in accordance with the Geneva Conventions’.30
However, in Mothers of Srebrenica, the Dutch courts established that there was sufficient evidence to confirm that, as of the evening of 12 July 1995, Dutchbat had known that the males separated ran a real risk of violation of their rights under Articles 2 and 3.31 Part 4.3.2 of the Supreme Court judgment provides a list of reasons suggesting that the Dutchbat command had both seen and received credible reports ‘that the Bosnian Serbs were subjecting the [Bosniac] male refugees to gross violence’.32 Between 100 and 400 males were murdered already in Potočari on 12 and 13 July 1995.33 Among others, Dutchbat had inspected the so-called ‘white house’,34 and seen the terrors of males kept there and that they had been divested of their identification en masse, indicating that executions awaited them. These findings further confirm the ICTY findings in Krstić, where the Tribunal held that Bosnian Serbs had subjected the people in the mini safe area to a terror campaign comprised of threats, insults, looting and burning of nearby houses, beatings, rapes and murders.35
In the same vein, the Dutch Supreme Court applied the adequate ECtHR test of ‘knew or ought to have known’. At the same time, one could argue that Dutchbat knew or ought to have known of the risks the males were facing at an earlier date, and at least by 11 July 1995. In recently unclassified documents pertaining to Srebrenica, it was made clear that, in the Netherlands Council of Ministers meeting held on 11 July 1995, a government minister had stated that ‘[t]he worst must … be feared for the men’. They mentioned, among others, the ‘worst fear’ and ‘ethnic cleansing’.36 Indeed, in the following days, the slaughter of Bosniac males commenced. While the Dutch Supreme Court applied the right ECtHR test of ‘the knew or ought to have known’, the information above raises the question that the Netherlands may already have known the risks the males were facing on 11 July 1995. Such knowledge, triggering the State’s positive obligations under Articles 2 and 3, is important also in the context of planning and operations to protect the lives of Bosniac males at least in the immediate aftermath of the fall of Srebrenica.
Articles 2 and 3 rank as the most fundamental provisions in the Convention. The right to life is a precondition for enjoyment of any other human right, and the prohibition of torture is an absolute right of jus cogens character. The ECtHR has held that Articles 2 and 3 must be construed strictly37 and applied so as to make their safeguards practical and effective.38 It has also held that deprivations of life must be subjected ‘to the most careful scrutiny’.39 In this connection, it is relevant that Article 2 requires the State ‘not only to refrain from the intentional and unlawful taking of life but also to take appropriate steps to safeguard the lives of those within its jurisdiction’.40 Such appropriate steps include the obligation on the State ‘to take preventive operational measures’.41 There are numerous obligations under Articles 2 and 3 ECHR. For the purpose of the Mothers of Srebrenica, however, two are most relevant: (i) the duty of care or means and (ii) the adequate decision-making, planning and operations in complex security situations. In what follows, I will first outline the ECtHR standards and then examine whether the Dutch Supreme Court diligently applied such standards in Mothers of Srebrenica.
The ECtHR is mindful that protection of all lives is not always possible and has acknowledged that
‘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, [the obligation to protect the right to life] … must be interpreted in a way which does not impose an impossible or disproportionate burden on the authorities’.42
At the same time, the ECtHR has clarified that the ‘authorities must take appropriate care to ensure that any risk to life is minimised’ and that authorities must be examined whether they ‘were not negligent in their choice of action’.43 The Strasbourg Court has noted the requirement on the State authorities to ‘undertake any measures within their powers that could reasonably be expected to avoid, or at least mitigate risk’.44 According to Sofia Galani
‘States simply have to adopt measures to avert an attack or minimize the loss of life caused by such an attack. The failure of these measures to achieve intended outcome does not breach Article 2 of the Convention. This shows that the obligation is one requiring due diligence, not an obligation of result’.45
Also, the Osman standard clarifies that the State must take measures that ‘might have been expected to avoid’46 and not that would have avoided the risk.
A State may violate Article 2 of the ECHR even if the victim is alive. A breach of the right to life does not depend on someone having died or survived. According to the ECtHR case-law, in certain circumstances,
‘depending on considerations such as the degree and type of force used and the nature of the injuries, use of force by State agents which does not result in death may disclose a violation of Article 2 of the Convention, if the behaviour of the State agents, by its very nature, puts the applicant’s life at serious risk even though the latter survives’.47
The Strasbourg Court has taken a similar approach in cases concerning persons who had survived a potentially lethal attack by other actors.48
The opposite rationale also applies. If an individual is killed by acts of violence of other actors, a State may not necessarily breach Article 2 of the Convention if it can demonstrate that it did all it could to protect or at the very least minimise the risk to life.
For the sake of argument, even if all Bosniac males had actually survived, there could still be a breach of the right to life. This simply illustrates that the essential issue under Article 2 of the ECHR in the present case concerns the duty of care and the obligation of a State to do all it can reasonably do to protect the right to life or at the very least minimise the risk to life. It concerns, among others, the behaviour of the State.
The Supreme Court in Mothers of Srebrenica did refer to the notion of duty of care.49 However, when unwrapping further its reasoning, it becomes clear that it misapplied the standard of duty of care as understood by the ECHR. The Dutch courts practically looked at whether Bosniac males would have had a chance of survival had Dutchbat done something differently, rather what Dutchbat had been expected to do to protect the human rights under Articles 2 and 3 of the ECHR. Two arguments from the Supreme Court’s reasoning are most telling in this regard. The first concerns the separation of males, and the second concerns the failure to report to the UN on the war crimes Dutchbat had been witnessing.
According to the lower Dutch courts, Dutchbat had ‘facilitated’ the separation of males by Bosnian Serbs and thus had violated these males’ rights under Articles 2 and 3 of the Convention.50 The Court of Appeal held that the separation of males was a ‘grave’ action.51 Even though the Court of Appeal could not establish that Bosniac males would have survived in Srebrenica had Dutchbat refrained from engaging in their separation, the act of facilitating the separation of the males was in itself in breach of Articles 2 and 3 of the ECHR.52 The Supreme Court dismissed these findings and ruled that, by cooperating with Bosnian Serbs in removal of people from the mini safe area, Dutchbat had acted lawfully. The Supreme Court instead held
‘As it was clear to Dutchbat that ceasing to cooperate would not affect the risk the male refugees outside of the compound were facing, continuing to cooperate in the evacuation was not wrongful under the circumstances of the case, taking into account the war situation’.53
The core aspect in the Supreme Court’s finding is causality and the assumption that if Dutchbat had ceased to cooperate with Bosnian Serbs, it ‘would not have avoided’ the killing of the males.54
Furthermore, it is noteworthy that the Supreme Court changed the focus of its assessment from separation of males to evacuation of females, and in that manner shifted the focus away from those who had actually been at risk when it held
‘Given the war situation in which decisions had to be taken under considerable pressure, and given the fact that decisions had to be taken based on a weighing of priorities, Dutchbat was reasonably entitled to opt to continue to cooperate in the evacuation by designating groups and forming a sluice, in order to – in any event – prevent chaos and accidents involving the most vulnerable people (women, children and elderly). Although the Court of Appeal has rightly ruled that, briefly put, the latter interest (obviously) carries “less weight” than the real risk the men were facing (para. 61.6 at d), this does not mean, contrary to the opinion of the Court of Appeal, that Dutchbat should have opted to disregard the interest of the women, children and elderly by ceasing to cooperate in the evacuation’.55
While focusing on the result and not the duty of means or care, this approach taken by the Supreme Court not only omits an essential standard of protection under ECHR but also appears to suggest that actions that ‘facilitate’56 the realisation by a third party of serious and large-scale human rights violations under Articles 2 and 3 can be compatible with the Convention.
As was already explained, as of at least 12 July 1995, Dutchbat knew of the risk to the lives of Bosniac males. Despite that, Dutchbat did not report to the UN the atrocities it was witnessing or that there were clear signs of mass killings to come.57 Despite this, the Supreme Court merely noted that, apart from the fact that nine dead bodies had been found on 13 July 1995, the Dutchbat had not reported any other war crimes.58 Earlier rulings held that, even if the Dutchbat’s failure to report war crimes as of 8 July 1995 had breached the Geneva Conventions, it could not be established that the reporting of war crimes would have led to such military intervention that could have saved the males’ lives.59
The rationale of the Supreme Court judgment is that, even if Dutchbat had reported war crimes to the UN, this would not have changed the fate of Bosniac males because the UN would not have intervened on time. The domestic courts found that since ‘the UN [had been] aware that the Bosnian Serbs [had advanced] on the safe area by force’,60 the reporting of atrocities would not have changed anything at the UN level in terms of its response. They also observed that military intervention had not been plausible within a few days because ‘after the war crimes had become public knowledge … on 10 August 1995, it took until 30 August 1995 for NATO … to carry out air strikes and bombings’.61 The domestic courts limited their assessment to one measure only: a military intervention.
The Dutch courts were right in finding that the UN knew that Bosnian Serbs were violent. Indeed, that was part of the reason if not the reason why peacekeeping mission was there in the first place. However, it is one thing to say that Bosnian Serbs advanced on the safe area and another to say that they are in a slaughter mission. Also, the Dutch courts were right in noting that the UN and NATO air support was not always prompt or even adequate. In such critical times, however, the information on the commencement of mass killings, judged reasonably, might have affected the UN response. One could argue that the UN and NATO might have been quicker in an intervention to prevent the genocide as opposed to the intervention in August 1995, after the genocide had taken place.
The failure to report the atrocities prevented any reassessment of decisions at the State or international level to plan and implement a response to mass atrocities. As long as States and international organisations lacked that essential information about the systematic killings and ill-treatment and, moreover, commencement of genocide, any conclusions on the likelihood of survival of Bosniac males may be viewed as arbitrary and a mere speculation.
In this connection, the UN report concluded that it was hard
‘to explain why Dutchbat personnel did not report more fully the scenes that were unfolding around them following the [Srebrenica] enclave’s fall … It is possible that if the members of the battalion had immediately reported in detail those sinister indications to the United Nations chain of command, the international community might have been compelled to respond more robustly and more quickly, and that some lives might have been saved’.62
Hence, the UN report, unlike the Dutch Supreme Court, finds that the detailed reporting of mass atrocities might have at the very least minimised the risk of killing and ill-treatment of the males.
Finally, it has not been established whether the Dutchbat’s failure to report mass atrocities was a political or operational decision. If it was the latter, this would also raise questions as regards the Dutchbat’s preparedness to serve in such complex security situations. The report on Srebrenica prepared by the Netherlands Institute for War Documentation (NIOD) stated that no reference could be found in the Dutchbat Standing Orders about what actions to take upon observing human rights violations63 and that after the fall of Srebrenica it had not occurred to Dutchbat command to gather information about the grave breaches of human rights.64 In that light, it was important for the Dutch courts to establish whether the failure to report mass atrocities had also been connected with the preparation of Dutchbat as required by Articles 2 and 3.65
The Dutch courts considered that the failure of Dutchbat to inform the UN and NATO about the commencement of genocide was lawful because, even if they had done so, that most likely would not have changed the fate of Bosniac males. The same rationale was applied in relation to separation of males outside the compound. The Dutchbat’s refusal to let all Bosniac males inside the compound was based on the same reasoning. This approach is the crux of the problem in the Mothers of Srebrenica case.
In general, these types of arguments led to a post factum speculation about the fate of the victims and failed to focus on what the State authorities had been required to do to minimise the risk to life on the critical dates of July 1995. If one is to speculate post factum on the chance of survival as a form of ‘legal reasoning’, endless creative arguments can be posed. One may speculate that, in fact, the chance of survival would have been above 60 % if Dutchbat had told all Bosniacs to run away to the woods because they were very likely to be killed otherwise. In fact, the majority of those who had fled to the woods appear to have survived. To add to this type of speculation, if Dutchbat knew that only males were at risk, they could have asked them, where possible, to switch clothes with females and try to look like females. This is the so called ‘wartime cross-dressers’ and has been practised over a long time. Mythology suggests that Achilles was dressed as a female to avoid the participation in the Trojan War.
The Dutch courts’ rationale of post factum assessment of chance of survival does not follow a strict standard of duty of care as required under Articles 2 and 3 of the ECHR. This led the Court to find there was no violation by the Dutchbat, despite the fact they committed acts assisting the wrongdoers (i.e. the separation of the males) and failed to take actions to minimise the risk to life (i.e. failing to report war crimes), simply because there was little to no chance of survival of victims. The focus on the result in the context of gross human rights violations undermines the raison d’etre of the Convention system. According to the ECtHR, the interpretation of Articles 2 and 3 must be consistent with ‘the general spirit of the Convention, an instrument designed to maintain and promote the ideals and values of a democratic society’.66 In this regard, the ECtHR has also been critical of any appearance of tolerance of unlawful acts by the public authorities.67 In the present case, this concerns not only the public confidence in an international peacekeeping force, but also a signal that Dutchbat cooperation might had sent to the wrongdoers.
One could argue that the Dutch courts’ outcome-based approach to the failure to report war crimes and separation of Bosniac males might have been due to the fact that Mothers of Srebrenica was decided in tort law proceedings. Tort law, however, may co-exist with but cannot lower the ECHR standards. This is true also for any other field of domestic law. Certainly, States Parties to the ECHR can in principle choose any legal framework to secure human rights. However, the Convention does not permit States to apply different ECHR standards because a national legal framework provides that certain type of cases are litigated as tort, criminal or administrative. If a national court is unable to reconcile a possible conflict between national law and ECHR standards, the latter must prevail. Indeed, in a different tort case, a Dutch court had found aspects of tort law regarding the award of non-pecuniary damages to be incompatible with ECHR standards, and has clarified that ECHR standards prevail in case of any such inconsistencies.68 Perhaps there is a need to reflect more broadly on how to reconcile tort and ECHR in the Dutch jurisprudence.
In addition to the duty of care, the ECtHR in its case-law has carefully assessed issues pertaining to decision-making, planning and operations to determine whether the actions or omissions by State authorities were (in)compatible with Articles 2 and 3 of the ECHR. In Tagayeva and others v Russia, the ECtHR looked at four guideposts in relation to planning and operations
‘(i) whether the operation was spontaneous or whether the authorities could have reflected on the situation and made specific preparations; (ii) whether the authorities were in a position to rely on some generally prepared emergency plan, not related to that particular crisis; (iii) … the degree of control of the situation …; and (iv) that the more predictable a hazard, the greater the obligation was to protect against it’.69
In Tagayeva and others v Russia, the ECtHR found Russia in breach of Article 2 for, among others, failing to prevent a terrorist attack. The case concerned the siege of a school in Beslan in 2004 by Chechen groups. Of 1,100 hostages, 800 had been children. The hostages had been surrounded by, among others, suicide bombers.70 Although Russia’s knowledge of the specifics of the threat in Beslan, the ECtHR relied on some intelligence and other information,71 and found that Russia had had a ‘relatively specific advance information’ regarding the imminence, nature and geographical location of the threat.72 The ECtHR found Russia in breach of Article 2 for its failure to take necessary operational measures and feasible precautions to minimise the loss of life.73 In addition, the ECtHR considered that the lack of coordinated response to the threat before the siege had amounted to breach of Article 2.74
By contrast, in the earlier case of Finogenov and others v Russia, to which the Supreme Court referred in Mothers of Srebrenica, the ECtHR looked at whether Russia’s planning and rescue operation in response to a terrorist hostage crisis had been compatible with Article 2 of the ECHR. The case concerned the siege of a Moscow theatre by around 50 Chechens, involving 850 hostages and resulting in around 170 deaths. The ECtHR considered that, although Russia had failed to take all feasible precautions to minimise the loss of life, this specific terrorist attack could not have been foreseen by Russia and thus it could not have been required to have had an emergency plan at the time.75 The important factual aspect in that case was that Russia had been taken by surprise and had had no knowledge of the terrorist attack underway.
Tagayeva, on the other hand, provides a clear standard that the preventive obligation requires a State to ‘undertake any measures within their powers that could reasonably be expected to avoid, or at least mitigate [the foreseeable] risk’.76 In assessing such measures adopted by the State, courts must have due regard of the decision-making, planning and operations of the State to establish whether the State did all it could to protect or at the very least minimise the loss of life. The assessment of such factors is based on the time when they were decided and planned and are not judged after the end of such events. Furthermore, the ECtHR has established that, when looking at the obligation to protect under Article 2, ‘the primary aim of the operation should be to protect lives from unlawful violence’.77 Hence, the ECtHR has established a victim-centred approach regarding the protection of life in challenging security situations.78
In Tagayeva, the ECtHR narrowed the margin of appreciation that it created in Finogenov and clarified that it would not ‘hesitate in scrutin[is]ing military and operational decisions’.79
It is puzzling that the Dutch Supreme Court in Mothers of Srebrenica referred to Finogenov but made no reference to Tagayeva. Certainly, it was reasonable for the Supreme Court to refer to Finogenov as Dutchbat operated in a very challenging security situation with limited control and resources.80 However, Dutchbat knew or ought to have known about the risk to the life of Bosniac males, and thus in some respect it resembled Tagayeva more than Finogenov.
As a minimum, the Supreme Court should have explained the adequacy of Finogenov and the non-application of Tagayeva. Certainly, the Supreme Court could have viewed that there is a middle ground between the two, or that Tagayeva does not apply. The lack of clarity in this regard leaves the impression that the Supreme Court was searching for a source (Finogenov) and not a standard (Tagayeva).
Be that as it may, when looking at the reasoning of the Dutch Supreme Court, it is difficult to see an explicit and careful assessment of the decision-making, planning and operations regarding the Dutchbat’s conduct in Srebrenica. Such assessment would require the examination by the courts of how decisions had been made and operations planned at the relevant time and not based on how events had unfolded and the results of the Dutchbat conduct at issue.
First, the information on whether there was an informed decision to cooperate with Bosnian Serbs in separation of Bosniac males remained unclear. The NIOD report recalls that different Dutch political and military leaders in the Netherlands and Srebrenica had a different recollection about whether there had been any ‘instruction … or a clear guideline from the Ministry of Defence’,81 and whether Dutchbat had reported accurate numbers of those who had sought refuge.82 The Supreme Court did not explain what the ultimate State decision on the separation of males had been, how it had been made or communicated. The omission in addressing this question leaves unanswered an important matter regarding decision-making, the lines of command and communication.83
Second, it remained unclear in Mothers of Srebrenica whether or how the State had made decisions with regard to resolving the various interests at stake, in particular the interest to ensure that Dutchbat troops returned home safely, and the interest to protect the Bosniacs on the ground. It appears that the Dutch Minister of Defence had noted on 10 July 1995 that ‘topmost priority’ should be given ‘to the safety of Dutch military personnel’ but also that the Dutchbat should avoid victims first and foremost.84 The failure to engage with the decision-making in this regard leaves unanswered the question of whether Mothers of Srebrenica followed the victims-centred approach of the ECtHR in relation to complex security situations or whether there was any consideration about the prioritisation of lives when deciding on how to respond to the Srebrenica crisis.85
Third, it would have been appropriate to establish how the decision not to inform Bosniacs of the risk to their lives was made. While the courts addressed this matter in relation to the approximately 350 males inside the compound, they did not assess whether there was also an obligation to inform the males outside the compound of the risk to their lives. The ECtHR has established that, when a State knows that certain individuals are at risk of life, a State must take ‘any steps with a view to informing [them] of the attack beforehand and to securing their evacuation’.86
Lastly, even if the situation in and around Srebrenica had been difficult, the Dutch courts did not examine whether also in such challenging circumstances there had been some means available to minimise the risk to life and not solely to avoid it. The Supreme Court noted in the facts that ‘[d]uring the attack [of 6 July 1995 by Bosnian Serbs] on the safe area the AbiH [namely, Bosniacs] asked Dutchbat repeatedly to be given (back) the arms handed in under the demilitarisation agreements’.87 One could argue that from the moment the UN troops were no longer able to secure the ‘safe area’, Dutchbat could have returned the arms it had collected in the 1994 demilitarisation process. However, the ‘Dutchbat denied these requests’.88 At the same time, these facts are not assessed in the context of the operational choices made by Dutchbat. The ECtHR has established in other cases that a State ‘must choose means and methods to avoid or minimise incidental loss of civilian life’.89 The ‘choice of means’ is scrutinised carefully by the ECtHR.90
Dutchbat operated in a very challenging situation. It has been well established that Bosnian Serbs had not been peaceful towards the UNPROFOR and, at times, had fired shots at and detained them. At the same time, Bosnia and Herzegovina was in the midst of war. Hence, it was no surprise that the place was hostile. In fact, it was the hostility that made it necessary to have armed contributing troops to the UN, including Dutchbat, in the country in the first place. The presence of the contributing troops, including Dutchbat, had also created an expectation in the local population that these troops would be willing and able to protect them. As was already mentioned, in 1994 Bosniacs had given away their arms, believing that in case of an attack by Bosnian Serbs they would be protected by those who had secured the demilitarisation process. It was also not surprising that, after the fall of Srebrenica, Bosniacs had fled to Dutchbat.
While there are strong reasons to empathise with the unfortunate position in which Dutchbat found itself, it cannot be argued that they should not have been prepared for such difficult situations. Hence, it was important for the Dutch courts to establish what existing (e.g. emergency) plan the Dutchbat had in 1995, and assess how they implemented that plan. As in Tagayeva, the Dutch courts should have examined, among others, ‘(i) whether the operation was spontaneous or whether the authorities could have reflected on the situation and made specific preparations; (ii) whether the authorities were in a position to rely on some generally prepared emergency plan, not related to that particular crisis’.91
The analysis above does not mean the Court would definitely have reached a different outcome. Nor does it intend to downplay the challenging security situation in which the Dutchbat found itself. Indeed, after carefully assessing the issues pertaining to decision-making, planning and operations, the Supreme Court may well have reached the same result or found no violation. The concern here lies with the standard of assessment used by the courts and not simply the outcome of the judgment.
Even though the Dutch courts established that, as of 12 July 1995, the conduct of Dutchbat was attributable to the Netherlands and not to the UN, this does not necessarily mean that they could not examine issues pertaining to the training, planning and operations prior to that date. In fact, the rather contested 12 July 1995 date can only serve as a starting time for the assessment of how Dutchbat implemented an existing plan and reflected on their operations in view of the situation on the ground.
In The Application of the European Convention on Human Rights to Military Operations, Stewart Wallace observes that, in light of the ECtHR case-law,
‘The positive obligation in Article 2 presents significant difficulties for States attempting to uphold their obligations during many military operations. It remains unclear what kind of margin of appreciation will be granted to the State and how flexible the Court will be in applying this obligation … As such the exact scope of the positive obligation remains unclear’.92
Indeed, the assessment of positive obligations in complex security situations is case-specific. One may argue that this is uncontroversial since a one-size-fits-all approach to protect life may not be suited to achieve the aim of protecting life. Different circumstances require different responses. At the same time, it cannot be argued that there are no guideposts in the ECtHR case-law when assessing positive obligations under the right to life in complex security situations. As explained above, the ECtHR has consistently examined how State authorities decided, planned and operated in situations where Articles 2 and 3 of the ECHR were triggered. The lack of an explicit and careful examination of decision-making, planning and operations raise prima facie issues of the standard of assessment in Mothers of Srebrenica.
The approach taken in Mothers of Srebrenica is important not only for the parties to the dispute, namely the Netherlands and the surviving relatives of the victims of Srebrenica genocide. Being a rare case in relation to genocide prevention in the post-World War II situation, Mothers of Srebrenica leaves an important legacy as regards tort law, immunity laws, State attribution and liability and human rights law. On the last point, the legacy is rather worrisome when it comes to the domestic courts’ reasoning on the duty of care as well as decision-making, planning and operations. As regards the duty of care or means, the courts’ outcome-based approach to the failure to report war crimes and separation of Bosniac males undermines not only the standard of the duty of care but also the spirit of Articles 2 and 3 of the ECHR, which ‘promote[s] the ideals and values of a democratic society’93 and is antagonistic to any appearance of tolerance by the public authorities of unlawful acts.94
The lack of explicit and diligent engagement with the details of decision-making, planning and operations have left many facts unaddressed about such essential issues as the separation of males and failure to report war crimes. While the Supreme Court could offer some discretionary power to the State authorities that operated in a very challenging security situation, it should have carefully assessed the exercise of discretion against the relevant ECHR standards under Articles 2 and 3. As Dworkin argues, ‘[d]iscretion, like the hole in a doughnut, does not exist except as an area left open by a surrounding belt of restriction … It always makes sense to ask, [d]iscretion under which standards?’95 It appears that the Supreme Court in Mothers of Srebrenica did not fully answer this question.
1Bosniacs are also known as Bosnian Muslims. The constitutional term Bosniac should not be confused with the term ‘Bosnians’ which denotes citizens of Bosnia and Herzegovina irrespective of their ethnic origin.
4ibid paras 28 and 39; Court of Appeal of The Hague 27 June 2017, ECLI:NL:GHDHA:2017:1761 (English translation ECLI:NL:GHDHA:2017:3376), paras 2.29, 2.40; District Court of The Hague 16 July 2014, ECLI:NL:RBDHA:2014:8562 (English translation ECLI:NL:RBDHA:2014:8748), para 2.34.
18Supreme Court 6 September 2013, ECLI:NL:HR:2013:BZ9225; Court of Appeal of The Hague 5 July 2011, ECLI:NL:GHSGR:2011:BR0133 (English translation ECLI:NL:GHSGR:2011:BR5388), 26 June 2012 ECLI:NL:GHSGR:2012:BW9015; Supreme Court 6 September 2013, ECLI:NL:HR:2013:BZ9228; Court of Appeal of The Hague 5 July 2011, ECLI:NL:GHSGR:2011:BR0132 (English translation ECLI:NL:GHSGR:2011:BR5386), 26 June 2012, ECLI:NL:GHSGR:2012:BW9014. See also Miša Zgonec-Rožej, ‘Netherlands v. Nuhanović Netherlands v. Mustafić-Mujić’ (2014) 108 American Journal of International Law 509.
19The Mothers of Srebrenica case was brought by the foundation Mothers of Srebrenica, a legal person established under the Dutch law with the aim to represent the surviving relatives of the Bosniac male victims. The Mothers of Srebrenica has litigated before Dutch courts from 2007–2019. In the first proceedings, they summoned the Netherlands and the UN regarding the failure to protect killings and torture in Srebrenica safe and mini-safe area. The Dutch courts maintained in all three instances that they did not have jurisdiction with regard to the UN because of their immunity. The second proceedings, from 2013 until 2019, were against the Dutch State only. See Supreme Court, judgment 19 July 2019, para 2.1.1.
21ibid para 5.1. For a detailed analysis of the judgment, see Cedric Ryngaert and Otto Spijkers, ‘The End of the Road: State Liability for Acts of UN Peacekeeping Contingents After the Dutch Supreme Court’s Judgment in Mothers of Srebrenica (2019)’ (2019) 66 Netherlands International Law Review 537.
22This submission does not address other matters related to the Mothers of Srebrenica case, such as attribution, immunities, fair trial, and the right to an effective remedy. They appear as separate contributions in this special issue.
36Hanneke Keultjes, ‘VN en Dutchbat hadden volgens minister meer kunnen doen om val Srebrenica te voorkomen’ Algemeen Dagblad 20 January 2021 <https://www.ad.nl/politiek/vn-en-dutchbat-hadden-volgens-minister-meer-kunnen-doen-om-val-srebrenica-te-voorkomen~a8dabc1a/> https://www.ad.nl/politiek/vn-en-dutchbat-hadden-volgens-minister-meer-kunnen-doen-om-val-srebrenica-te-voorkomen~a8dabc1a/ last accessed on 7 February 2021.
47Guide on Article 2 of the European Convention on Human Rights, 31 December 2020, 6–7. https://www.echr.coe.int/Documents/Guide_Art_2_ENG.pdf. Last accessed 7 February 2021.
57In this connection, see UN General Assembly, Fifty Fourth Session, A54/549, Report of the Secretary-General, The fall of Srebrenica, 15 November 1999, paras 358, 389–390. See Court of Appeal 2017 (n 4) para 2.54; District Court 2014 (n 4) para 4.235.
63The State Institute for War Documentation (RIOD) (succeeded by NIOD), Srebrenica: Reconstruction, background, consequences and analyses of the fall of a ‘safe’ area, 10 April 2002, pp. 2102–2103 (English-language version of the report) (the ‘NIOD report’). https://www.niod.nl/en/srebrenica-report/report. Last accessed 7 February 2021.
65As regards the responsibility of a State to ensure adequate training, see Sašo Gorgiev v the former Yugoslav Republic of Macedonia (ECtHR, 19 April 2012); Şimşek and Others v Turkey (ECtHR, 26 July 2005) para 109.
68District Court of The Hague 7 October 2020, ECLI:NL:RBDHA:2020:10058, 3.25. See also in this Special Issue Rianka Rijnhout, ‘Mothers of Srebrenica: causation and partial liability under Dutch tort law’ (2021) 36(2) Utrecht Journal of International and European Law.
The author is grateful to Zane Ratniece, Cedric Ryngaert and the anonymous reviewers for their helpful comments on earlier drafts. This paper is written in a personal capacity, and the views of the author do not necessarily represent the views of any entity with which he is affiliated.
Kushtrim Istrefi provided legal advice in Stichting Mothers of Srebrenica v. the Netherlands and Subašić and Others v. the Netherlands (2020) concerning the domestic proceedings in the Mothers of Srebrenica case.
Kushtrim Istrefi is Assistant Professor of Human Rights Law and International Security Law at Utrecht University. He is part of the Netherlands Institute of Human Rights (SIM) and the Utrecht Centre for Accountability and Liability Law (Ucall).