Search and Rescue Operations at Sea: Who is in Charge? Who is Responsible?

 Posted by: Trevisanut Seline

Symposium on the Law of the Sea and the Law of Responsibility, cross-posted on Opinio Juris

On Sunday, 8 May 2011, the British newspaper The Guardian reported the story of a boat carrying 72 persons, among them asylum seekers, women and children, which left Tripoli (Libya) for the Italian island of Lampedusa at the end of March 2011 (for comments, see here and here). After 16 days at sea, the boat was washed up on the Libyan shore with only 11 survivors. During the 16 days route, survivors told that they used their satellite phone, which later ran out of battery, to call an Eritrean priest in Rome for help (see Resolution 1872 of the Parliamentary Assembly of the Council of Europe). The priest alerted the Italian Maritime Regional Coordination Centre, which located the migrants’ vessel and sent out many calls to the ships in the area. Pursuant to survivors’ testimonies, on about the tenth day of their voyage, when half of the passengers were dead, a large aircraft carrier or helicopter-carrying vessel (probably involved in the NATO’s Operation Unified Protector, which was on going at that time off the Libyan shores) sailed near to the boat, close enough for the survivors to see the sailors on board looking at them with binoculars and taking photos. But no one rescued them.

Flag states and coastal states have a duty to render assistance to persons found at sea in the danger of being lost and people in distress (Article 98 of the United Nations Convention on the Law of the Sea (UNCLOS)). This core obligation under both treaty law (see also the 1974 Convention on the Safety of Life at Sea (SOLAS Convention), the 1979 Search and Rescue Convention (SAR Convention) and the 1989 International Convention on Salvage) and customary law applies in any maritime zone and in relation to any activity there performed. While implementing this duty states can either perform directly the search and rescue (SAR) operations, namely through their own SAR services, or ask a vessel, which is located in the proximity of the endangered persons, be it any merchant ship or the state vessel of another country, to perform the rescue operation.

The texts here mentioned expressly refers to states, flag or coastal. Practice however offers more and more examples of police activities performed under the command of an international organization or a supranational body. A question then rises:

• who are the bearers of the obligation? Namely, are those ‘entities’ (such as NATO) bound by the duty to render assistance?

The answer could be affirmative only if we consider that the duty to render assistance under customary international law has a wider scope of application ratione personae, if compared with the same obligation under treaty law. The practice however does not allow yet such a conclusion.

Another question raised by recent practice concerns the recipient of the obligation:

• Is the duty to render assistance a purely inter-state obligation or does it entail a right to be rescued for people in distress at sea?

The law of the sea allocates obligations and rights in different maritime zones among states. However, the multiplication of activities at sea and the increased human presence lead to the question of the protection of the human element, in particular of the application of human rights at sea. Many scholars have already discussed the application of the relevant human rights treaties at sea, pointing out that the law of the sea, specifically UNCLOS, pursues some community interests, among which the protection of human rights (see a.o. here). Some authors have stressed the complementarity of these two fields of international law (see here). The duty to render assistance can then be considered as the corollary obligation of the right to life when applied at sea and, consequently, a right to be rescued for the people in distress at sea can be deduced from the application of human rights instruments at sea.

Pursuant to the case law of the European Court of Human Rights (ECtHR), when the SAR operation consists in taking on board of the rescuing vessel the persons in distress, the latters fall within the jurisdiction of the flag state and the European Convention on Human Rights applies (Hirsi case). This is perfectly in line UNCLOS Article 92, which affirms the exclusive jurisdiction of the flag state on its own vessels on the high seas. The determination of the exercise of jurisdiction might however result more difficult when the rescued persons are not taken on board of the rescuing unit, or when the SAR services are not performed.

In the ECtHR Xhavara case, which did not concern a SAR operation, the ECtHR recognized the exercise of jurisdiction by Italian authorities on two grounds: the activities were performed on the basis of a bilateral agreement; and the fact that there was a collision between the Italian navy vessel and the migrants’ boat. Applying this reasoning by analogy, when the SAR operation is performed on the basis of an international agreement and there is contact between the rescuing unit and the vessel in distress, the ECHR applies.

The non-performance of SAR services is a trickier instance. The main question to be answered here is whether human rights instruments do apply in cases of SAR non-performance, i.e. whether the non-rescued persons do fall within the jurisdiction of any state while they are in distress at sea. It can be argued that the distress call creates a ‘relation’ between the state, which receives it, and the persons who send it. The life of the persons in distress depends on the behaviour of the recipient state. The argument could go further and support the existence of an exclusive long distance de facto control that the state, which received the call, exercises on the lives of those people. Their lives are submitted to the discretion of that state, in the case mentioned above the lives of the migrants depended on the Italian authorities and there capacity to activate SAR services.

This de facto control becomes also de jure when the distress situation is located within the SAR zone of the recipient state, which has an obligation to ‘promote the establishment, operation and maintenance of an adequate and effective search and rescue service’ (UNCLOS Article 98(2), emphasis added). There is then an assumption of authority (ECtHR Al-Skeini and others case) the coastal state exercises in its SAR zone. The coastal state is then bound by an obligation of due diligence. The loss of lives or the risk of losses in its SAR zone may entail a violation of the right to life of the concerned persons, of their ‘right to survival’, and a violation of its obligations under law of the sea instruments.

Update – The Dutch Courts and Asylum at the ICC: From Shared Obligations to Obligations of No One

 Posted by: Irving Emma

In October 2012, the SHARES Blog carried a post that discussed a September Dutch Court decision concerning the on-going asylum situation at the ICC. Since then there are have been two important developments: the matter has been taken to the European Court of Human Rights (ECtHR) and the original Dutch decision has been overturned on appeal.

First, one of the detained witnesses sought to challenge his detention by bringing a claim against the host State of the ICC before the European Court of Human Rights (ECtHR). However, the application in Longa v The Netherlands was declared inadmissible on the basis that The Netherlands had no jurisdiction under Article 1 of the European Convention on Human Rights. While acknowledging that jurisdiction is primarily linked to territory, the ECtHR drew on previous case law dealing with the ICTY to decide that the mere presence of the ICC on Dutch territory is not sufficient to engage Dutch jurisdiction. It went on to dismiss the arguments made that the level of human rights protection at the ICC is deficient, and then rejected the notion that by accepting jurisdiction to hear the asylum claims the Netherlands had also accepted jurisdiction under the Convention. The Court held that there was no obligation on member States of the ECHR to allow a person to await to outcome of an asylum claim on its territory.

Armed with the Longa decision, the Dutch authorities filed an appeal to the earlier Dutch Court decision that would have compelled The Netherlands to take over control of the witnesses. The Appeals Court overturned the first instance decision on 18 December 2012. In this decision, the Dutch Appeals Court denied the lower court’s assertion that the witnesses were in a dead end detention situation, which as it had no end in sight, had therefore become illegal. It was held that just because the asylum procedure was lengthy, did not mean it was unending. The Court held that in any case, it was not for it to consider whether there was a conflict between the detention and Articles 5 and 13 ECHR: as the ECtHR had said in Longa, the Netherlands lacked jurisdiction under Article 1.

Once the ECtHR handed down its opinion on the situation of the detained witnesses, The Netherlands was offered an easy way out of the compromising position which the judge of the first instance court had placed it in. The Dutch authorities were no longer compelled to take over the witnesses, and suffer the precedential and political consequences that would entail. However that does not remove the question of whether the Longa decision was indeed a good one. On closer inspection, one might wonder whether the precedent the ECtHR relied on was not easily distinguishable. Having read the arguments of the applicant, one is struck by the fact that the ECtHR seems to have overlooked a number of relevant facts which may have led it to a different conclusion.

It is to be regretted that an unconvincing decision of the ECtHR has left the witnesses detained at the ICC unprotected. The September Dutch Court decision arguably took the better approach to the legal questions raised by the situation. Now the most pertinent question is to whom can the witnesses turn to seek relief? Not to the ICC, nor the DRC, and now not to the Netherlands. The result is an example of where shared obligations result in obligations for no one.

Sharing responsibility? The co-respondent mechanism and EU accession to the ECHR

 Posted by: Lock Tobias

The negotiations on accession by the EU to the ECHR have recently entered their final stage. Representatives of the EU and of the forty-seven parties to the ECHR (the so-called Group 47+1, which has published the reports of its first and second meeting) are scheduled to meet twice this autumn to agree on a final accession agreement. It is likely that it will largely be based on the draft agreement on EU accession, which was made publicly available last year. Once final agreement has been reached, it is very likely that the Court of Justice of the EU (CJEU) will be asked for an Opinion on whether the agreement is compatible with the EU Treaties. Should the CJEU give the agreement a green light, the ratification process can begin.

This blog entry asks whether the co-respondent mechanism provided for in the draft agreement leads to a sensible allocation of responsibility between the EU and its Member States for violations of the Convention post-accession.[1] By introducing the co-respondent mechanism, the drafters aimed to resolve the problem that after accession both the EU and its Member States will be parties to the ECHR. Since it is mainly the EU’s Member States which implement obligations laid down in EU primary and secondary law, an applicant will usually hold the implementing Member State responsible in the European Court of Human Rights after having exhausted the domestic remedies in that Member State’s legal order. The problem is that the violation of human rights may not necessarily be the Member State’s fault as it may well have been contained in a provision of EU law. Thus even if the Member State is convicted in Strasbourg, it will be unable to remedy the violation since it lacks the competence under EU law to do so. In such cases it would therefore make sense to involve the EU in proceedings and thus to extend the binding reach of the judgment to it. The same is true in the reverse situation where the EU is held responsible for alleged violations contained in its primary law (mainly the EU’s Treaties). In such a case it would make sense to make the Member States join the proceedings as it is only they which can agree on a Treaty change.

As a solution to this problem, Article 3 of the draft accession agreement introduces the co-respondent mechanism. It provides that where a Member State is taken to the ECtHR and it appears that the alleged violation of the ECHR calls into question the compatibility with Convention rights of a provision of EU law, the EU may become co-respondent. The same is true in the reverse situation. Where the EU is respondent in proceedings before the ECHR and it appears that the allegation calls into question the compatibility with Convention rights of a provision of EU primary law, the Member States may become co-respondents.

The status of co-respondent will be unique in the ECHR legal order. Its main features are that the co-respondent becomes a party to the proceedings so that the judgment is also binding on it. The key difference to the status of an ordinary respondent is that the applicant does not need to exhaust the domestic remedies in the co-respondent’s legal order. Procedurally, it therefore reduces the burden for the applicant and is to be welcomed.

The main flaw of the mechanism, as laid down in the draft agreement, is that the co-respondent cannot be forced to join proceedings. Rather, the status is voluntary. This calls into question the main rationale behind the mechanism, which is the ‘correct’ attribution of responsibility between the EU and its Member States. From the perspective of EU law, this attribution cannot be undertaken by the European Court of Human Rights since it would violate the autonomy of European Union law if a court other than the EU’s Court of Justice made such an assessment (for details see here). The rationale for the voluntariness of the co-respondent status given in the explanatory report is that a party cannot be forced into proceedings where it was not named in the initial application. Yet this argument is based on a mere technicality and is hardly convincing. There is no reason why an applicant should not ask another party to join proceedings at a later stage provided that the procedural rights of that party are respected, especially regarding their right to be heard. In addition, the draft appears to contradict the rationale for the co-respondent mechanism. According to the explanatory report, the main reason for its introduction is to avoid gaps in the participation, accountability and enforceability in the Convention system. But a voluntary mechanism leads to exactly such gaps. The EU or a Member State may well decide to steer clear of proceedings even though they would be responsible for the alleged violation. The fact that the applicant can nominate both from the outset does not remedy this problem since it would force the applicant to exhaust domestic remedies in all legal orders concerned, which is not only costly but also time-consuming. It should especially be borne in mind that an applicant before the ECtHR does not need to be represented by a lawyer so that she may not even be remotely aware of these requirements. It follows that the current draft only achieves sub-optimal results regarding the efficiency of human rights protection for individuals.

This is coupled with the danger that a Member State raises the defence that it was not responsible for the violation as the violation was rooted in its obligations under EU law and it had no discretion. If such a defence were possible, neither the EU nor the Member State would be responsible in the ECtHR[2], which would defeat the whole purpose of EU accession. The draft and the explanatory report are silent regarding this possibility. It is important that the ECtHR would not accept such a defence. Otherwise, the human rights protection as regards EU law would risk being less effective than it is now. It is suggested that the draft presupposes that no such defence should be possible. It is therefore suggested that the responsibility in such situations should generally be joint. Neither the Member State nor the EU should have a possibility to escape it. This seems to have been recognised by the EU as it tabled an amendment to the draft accession agreement during the first meeting of Group 47+1, which states that acts of the EU’s organs remain attributable to the Member States even where they occur when the Member States are implementing their EU law obligations.[3]

It would therefore be welcome if the on-going negotiations on EU accession revisited this issue and came to a more convincing solution. The co-respondent mechanism is generally a good solution to balance the interest of the EU and its Member States in not allowing an external court to decide who was internally responsible with the interest of a potential victim of a human rights violation to get an effective remedy. But the co-respondent mechanism can only work if it leads to truly shared responsibility before the ECtHR. Otherwise, there is a risk of thwarting the overall aim of EU accession to the ECHR, which is the improvement of human rights protection in Europe.

For a more detailed version of the arguments made in this blog post see my forthcoming article “End of an epic? The draft agreement on the EU’s accession to the ECHR” in the Yearbook of European Law 2012. A pre-edited version of that article can be found here.

---

[1] This blog post repeats part of the argument I made at my SHARES lecture on 4 October on EU accession to the ECHR and the co-respondent mechanism. It is limited to explaining the main rationale for the co-respondent mechanism and to criticising its voluntary character. Other issues of shared responsibility could not be addressed (e.g. the question of who should have to pay ‘just satisfaction’ where the co-respondent mechanism applies).

[2] This argument assumes that in most cases an applicant would not be able to bring a case against EU legislative acts directly against the EU since it would lack the status of a victim required by Article 34 ECHR.

[3] Meeting report 47+1(2012)R02, page 15.