I. INTRODUCTION: WHAT TO DO WITH A CAPTURED PIRATE?
Piracy in the Gulf of Aden and off Somalia's east coast has become endemic. Sophisticated and aggressive pirates operating from mother-ships now board vessels hundreds of nautical miles from land, seeking to capture and ransom both vessel and crew. As at 11 November 2009, at least 13 vessels and 230 crew were being held to ransom off Somalia.Footnote 1
Since August 2008, the international response has been to send warships to the Gulf of Aden to protect shipping. There are now on average over 20 warships in the region at any time. A series of Security Council Resolutions commencing with Resolution 1816 in June 2008 have encouraged high seas counter-piracy patrols and authorized operations within Somalia's territorial sea and on its soil. In practice, naval forces initially focused on ‘deter and disrupt’ or ‘catch and release’ strategies. The former used warships as both a passive deterrent and active disrupter of pirate attacks. ‘Catch and release’ went further: suspect pirate vessels were boarded, weapons and boarding equipment thrown overboard, biometric data gathered from suspects and the vessels left with enough food and fuel to return to shore. These strategies had sound operational rationales, as taking a pirate into port for prosecution or transfer back to the seizing warship's flag State could involve a lengthy diversion of resources from protecting other vessels. However, two factors influenced a change in this approach. First, ‘catch and release’ proved politically unpopular, with politicians increasingly calling for an end to pirate ‘impunity’.Footnote 2 Second, situations emerged that ‘catch and release’ did not contemplate, such as pirates hijacking artisanal fishing vessels to use as mother-ships and keeping fishermen prisoner aboard. On 12 November 2008 boarding craft from the HMS Cumberland subdued a suspect pirate vessel and onboard Royal Marines discovered Yemeni fisherman being held by Somali pirates. They could scarcely all be released together, so something had to be done with the pirates. The pirates were soon transferred to Kenya for trial under an ad hoc agreement. This was not unprecedented; the US similarly transferred suspect pirates to Kenya in 2006.Footnote 3
The Cumberland and similar incidents revealed the complexities of counter-piracy law-enforcement. Where will pirates be tried? What are the standards for evidence in that forum? As one official remarked, ‘before you go looking for pirates, it's a good idea to be clear about who will be assuming jurisdiction’.Footnote 4 Few States apparently thought this through in advance; those that did often encountered national legal problems.Footnote 5 In 2008 the Danish warship Absalon had to set seized piracy suspects ashore after six days looking unsuccessfully for a means of prosecuting them.Footnote 6
Great efforts since have been directed at getting piracy suspects to trial. At the time of writing, Somali piracy suspects are being tried before courts in Somalia (Puntland), Kenya, Yemen, France, the Netherlands, and the US. Prosecution in a regional State, however, remains the preferred solution and Kenya is the venue of choice. Kenya has concluded arrangements governing transfers of piracy suspects for trial with the US, the UK, Denmark and the EU.Footnote 7 Regional States also concluded a non-binding code of conduct on combating piracy in January 2009, discussed further below. Relevant Security Council Resolutions have encouraged these and other efforts, including the establishment of the Contact Group on Piracy off the Coast of Somalia (a grouping of 45 governments and various international organizations) which is committed to taking further legal, political and practical steps to counter Somali piracy.Footnote 8
Thus, in the remarkably short period from April 2007 to mid-2009 a flurry of hard- and soft-law-making activity has constructed an increasingly comprehensive and decentralized legal framework addressing piracy off Somalia. Warships operate under a variety of national and international mandates, and there is no single unified command covering all counter-piracy missions. Bilateral transfer agreements sit alongside a regional code of conduct, and both sit within the framework of the law of the sea and Security Council Resolutions. One-size-fits-all solutions have been eschewed for a pragmatic range of national and international mechanisms.
Nonetheless, captured piracy suspects raise significant questions about their arrest, detention and transfer to a State willing to try them, especially for State parties to the European Convention on Human Rights. The present article will first outline the general legal framework surrounding Somali piracy (parts II and III) before returning to specific human rights issues (part IV) and potential ways forward (part V).
II. THE INTERNATIONAL LAW OF PIRACY
The international law applicable to piracy is straight-forward, if bedevilled by historical confusion. The real difficulty is the interface with (and lack of appropriate) national laws. International piracy law is largely codified in articles 101–107 and 110 of the UN Law of the Sea Convention (UNCLOS) and articles 14–22 of the High Seas Convention. These define piracy as:Footnote 9
(1) ‘any illegal acts of violence or detention, or any act of depredation’;Footnote 10
(2) committed for private ends;
(3) on the high seas or in a place outside the jurisdiction of any State; and
(4) committed by the crew or passengers of a private craft, against another vessel or persons or property aboard.
While now generally accepted as customary law, there are worse and better arguments that this definition is not, historically, a codification. Worse arguments focus on national decisions of the 19th or early 20th century proclaiming acts in a foreign State's territorial sea or aboard a single vessel may be piracy by ‘the law of nations’.Footnote 11 These cannot be a reliable guide to the law a century later, not least because such cases notoriously conflated elements of national and international offences.Footnote 12 Better arguments note the diverse, contradictory historical case law and scholarly commentary and ask how any coherent rule could emerge from such incoherent material.Footnote 13 Both arguments fail to acknowledge that the successive re-enactment of this definition in treaties and regional instruments evidence States' present acceptance of it as customFootnote 14 no matter how unconvincing this proposition might seem viewed over the longue durée.
Once this definition is accepted, several inherent limitations should be noted. First, piracy must be committed from one vessel against another. Internal hijackings, as occurred in the 1961 Santa Maria or 1982 Achille Lauro incidents,Footnote 15 are not piracy of themselves. This presents no problem regarding Somali piracy, where pirates attack their targets from small boats.
Second, it is commonly held that the ‘private ends’ requirement means politically-motivated acts cannot be piracy. The better view is that any act of violence not sanctioned by State authority is one for ‘private ends’, the correct dichotomy being not ‘private/political’ but ‘private/public’.Footnote 16 Regardless, not only are Somali pirates' ransoms demands clearly sought ‘for private ends’, but on some reports pirates have even announced to their captives that they are acting ‘for private ends’.Footnote 17 This is commercially sound: if Somali pirates were considered terrorists ransoms might not be payable under financing of terrorism laws.
Third, a geographic limitation governs both the offence itself (as a matter of prescriptive jurisdiction) and certain of the powers granted to suppress it (enforcement jurisdiction). Piracy can only be committed on the high seas, being all waters beyond the territorial sea, including the EEZ.Footnote 18 Equivalent acts within territorial or internal waters are a matter for the coastal State's criminal jurisdictionFootnote 19 and are now commonly termed ‘armed robbery against ships’.Footnote 20 On the high seas, any government vessel may board a vessel suspected of piracy as an exception to the otherwise exclusive jurisdiction of the flag State.Footnote 21 Where evidence of piracy is discovered the State vessel may seize the suspect vessel, arrest persons on board, and subject them to the jurisdiction of the seizing State's courts.Footnote 22 By definition, these powers are granted only on the high seas and do not extend to pursuing pirates into foreign territorial waters.Footnote 23
Customary international law permits any State subsequently finding a pirate within its territory to prosecute him or her as an exercise of universal jurisdiction.Footnote 24 The mere existence of such jurisdiction does not necessarily oblige States to use it. While UNCLOS requires that States must ‘cooperate to the fullest possible extent in the repression of piracy’ (article 100), it only provides that a seizing warship may send pirates for trial before national courts (article 105).Footnote 25 The inference is that States have no duty to enact relevant national offences and have ‘a certain latitude’ to cooperate in suppressing piracy by means other than prosecution.Footnote 26
A further gap between articles 100 and 105 is that no express rule governs the transfer of suspected pirates from a seizing State to a nearby port State. Nonetheless, any State could accept such a transfer and assert universal jurisdiction over such received suspects. It has been argued against this that UNCLOS article 105 refers only to the seizing State's adjudicative/curial jurisdiction.Footnote 27 While correct, this does not preclude the existence of other valid jurisdictions nor prevent transfers between them. While the ILC commentary considered that this right ‘cannot be exercised at a place under the jurisdiction of another State’,Footnote 28 this only suggests a seizing State cannot exercize its own judicial powers in foreign territory.Footnote 29 Nothing precludes a ‘receiving’ State exercising its own independent jurisdiction. Indeed, all relevant State practice supports such a power of transfer, though not a duty to accept transfers. This raises questions of burden-sharing. Which State, absent agreement, should be responsible for trying a suspect pirate: the seizing State, the pirate's State of nationality, a nearby port State or the State with ‘strongest links’ to the offence? International law provides no rule of priority between competing potential jurisdictions,Footnote 30 let alone one of obligation.Footnote 31 Burden-sharing is thus a political, not a legal issue.Footnote 32
The question is, of course, the extent to which this general law has been modified regarding Somalia by relevant Security Council Resolutions. These are considered below.
III. THE SECURITY COUNCIL AND SOMALI PIRACY
A. Origins of the Counter-Piracy Resolutions
International Maritime Organization (IMO) statistics show a leap in attempted and successful pirate attacks off East Africa after 2004. By 2007 more high-seas attacks were reported off East Africa than for the traditional piracy hot-spots of the Malacca Strait and South China Sea combined.Footnote 33 The jump is attributable to Somali pirates' transition from ‘disorganized’ smash-and-grab or stick-up piracy (where pirates seize cash and valuables and then leave the vessel) to highly organised hostage-takings throughout 2000–2005.Footnote 34 Once the success of this new business model was established, it was widely adopted. What began as the alleged self-help remedy of displaced fishermenFootnote 35 is now highly-developed organized crime, worth US$18–30 million in 2008 alone.Footnote 36
The IMO requested that Somalia's Transitional Federal Government (TFG) and the Security Council take urgent action on piracy in December 2007.Footnote 37 Somalia advised the Security Council of its consent to such measures on 27 February 2008. Despite the Security Council expressing concern about Somali piracy as early as 2006,Footnote 38 no substantive measures were forthcoming (other than in protection of World Food Programme convoys)Footnote 39 until France sponsored a resolution following the Le Ponant hijacking of April 2008.Footnote 40
On 4 April 2008 crew and passengers aboard the French yacht Le Ponant, including 22 French nationals, were taken hostage by pirates in international waters off Somalia. A week later the hostages were released at sea following a ransom payment; French helicopters and commandos then seized their captors on shore.Footnote 41 France had sought and received the TFG President's permission on 5 April 2008 for this operation.Footnote 42 The pirates were subsequently removed to face trial in Paris with the TFG's separate permission.Footnote 43 The French Prime Minister declared on 12 April 2008 his hope to create a UN-mandated international counter-piracy force to patrol waters off Somalia and other areas such as the Gulf of Guinea and the Malacca Strait.Footnote 44 Later in April, France, with the US and Panama, introduced a draft text before the Security Council,Footnote 45 subsequently adopted as UNSC Resolution 1816. This Resolution and its successors have only partially realized France's stated ambitions. While the Resolutions have only limited impact on the transfer of pirates to regional States, they nonetheless provide the backdrop to counter-piracy operations.
B. General Features of the Counter-Piracy Resolutions
The relevant counter-piracy Security Council Resolutions are 1816, 1846, 1851, 1897 and to some extent 1814 and 1838. Of the latter two, Resolution 1814 principally concerns protection of humanitarian aid deliveries to SomaliaFootnote 46 and requires no detailed discussion here. Resolution 1838 concerns measures on the high seas generally, and is discussed below. Resolutions 1816, 1846 and 1851 concern, in part, actions in Somalia's territorial waters or land territory. Each contains the talismanic Chapter VII authority to use ‘all necessary means’ to counter piracy. These three ‘territorial’ Resolutions contain a grant of power and a procedural limitation. Resolutions 1816 and 1846 grant authority to ‘cooperating States’ to enter Somalia's territorial sea to repress piracy in a manner consistent with the international law applicable on the high seas. This might involve a kind of reverse ‘hot pursuit’ of pirates into, or counter-piracy patrols within, Somalia's territorial waters. Generally speaking, this has not been the subsequent practice. Most, if not all, operations against Somali pirates occur on the high seas. Resolution 1816, initially to operate only for six months, was renewed by Resolution 1846 on 2 December 2008.Footnote 47 Resolution 1816 was discussed here in 2008 and that discussion remains relevant to Resolution 1846.Footnote 48 Resolution 1851 authorises ‘cooperating States’ to go further and engage in counter-piracy action on Somali soil.Footnote 49 To be a ‘cooperating State’ under any of these resolutions a State must be operating with the consent of the TFG as notified in advance to the UN Secretary General.Footnote 50 This provision makes the resolutions prima facie superfluous. Chapter VII is not needed to permit consensual operations; Somalia may authorize foreign law-enforcement action within its jurisdiction as in the Le Ponant affair. Nonetheless, these Resolutions serve a purpose. Some States find it politically or constitutionally easier to commit military forces to UN mandated extraterritorial operations. Others may doubt the TFG's ability to give timely authorization. Having a system for granting authorization in advance is certainly faster than negotiating ad hoc with high-ranking officials under pressure of time. Crucially, States not having formally recognized the TFG can rely on the Resolutions, rather than direct TFG consent.Footnote 51 The authority granted in Resolutions 1846 and 1851 was set to expire in December 2009 but on 30 November 2009 Resolution 1897 effected the extension of the relevant provisions for a further 12 months.
Resolution 1838 ‘[c]alls upon States whose naval vessels … operate on the high seas … off … Somalia to use … the necessary means, in conformity with international law, as reflected in [UNCLOS]’ to repress piracy.Footnote 52 ‘Necessary means’ is ordinarily interpreted as authorising military force. Here ‘means’ is limited to actions ‘in conformity with international law’, and the Security Council has repeatedly indicated that UNCLOS codifies the relevant law.Footnote 53 The result, then, is to grant no powers not already existing at international law. Similarly, Resolutions 1816 and 1846 simply encourage States to ‘coordinate their efforts to deter … piracy’ acting ‘in accordance with relevant international law’.Footnote 54 These Resolutions, then, are largely hortatory and grant no novel high-seas powers, subject to one possible exception.
Resolutions 1846, 1851 and 1897 all call upon ‘States, regional and international organizations … to take part actively in the fight against piracy’ including by ‘seizure and disposition of boats, vessels, arms and other related equipment used [or where there are grounds for suspecting such use] in … piracy … off the coast of Somalia’.Footnote 55 This may grant a power to summarily dispose of equipment suspected of being intended for use in piracy not available under UNCLOS. The express reference to ‘international organizations’ also clarifies the position of NATO and the EU. Finally, in respect of land operations, Resolution 1851 requires States to comply with ‘applicable international humanitarian … law’ (‘IHL’). Some have suggested that this makes all of IHL applicable to counter-piracy operations on Somali soil.Footnote 56 This is clearly wrong: the resolution refers to ‘applicable’ IHL, i.e. law that would apply in any event. Pirates are criminals to be captured using reasonable force; not combatants who may be targeted with lethal force or civilians immune from direct targeting under IHL.Footnote 57 Issues regarding the use of force against pirates are, however, beyond the scope of this article.Footnote 58
The Resolutions also all emphasize, in a relatively standard savings clause, that they were passed with Somalia's consent, that they apply only to Somali piracy, and do not serve as a precedent for customary law or modify parties' rights and obligations under UNCLOS.Footnote 59
C. From Military Mandate to Law Enforcement
The counter-piracy Security Council Resolutions exhibit a shifting set of concerns between ‘top-down’ military mandates and ‘bottom-up’ law enforcement cooperation. In June 2008, Resolution 1816 simply urged all States ‘to cooperate with each other, … the IMO and … relevant regional organizations in connection with … acts of piracy and armed robbery … off the coast of Somalia’.Footnote 60 This language was repeated on 2 December 2008 in Resolution 1846,Footnote 61 but with the addition that the Council:
Notes that the 1988 Convention for the Suppression of Unlawful Acts Against the Safety of Maritime Navigation (‘SUA Convention’) provides for parties to create criminal offences, establish jurisdiction, and accept delivery of persons responsible for or suspected of seizing or exercising control over a ship by force or threat thereof … ; urges States parties to the SUA Convention to fully implement their obligations … and cooperate with the Secretary-General and the IMO to build judicial capacity for the successful prosecution of persons suspected of piracy and armed robbery … off the coast of Somalia …Footnote 62
Other than the reference to judicial capacity-building, this paragraph is aimed at two of the problems previously identified. First, international law does not require a port State to receive suspected pirate from the seizing vessel. Second, States are not obliged to have national laws criminalizing piracy. Resolution 1846 looks to the SUA Convention to bridge both gaps. Somali hostage-taking piracy clearly falls within the SUA Convention offence of ‘seiz[ing] or exercis[ing] control over a ship by force’ when navigating in international waters.Footnote 63 Article 8(1) of the SUA Convention provides that the ‘master of a ship of a State Party (the “flag State”) may deliver to the authorities of any other State Party (the “receiving State”) any person who he has reasonable grounds to believe has committed … [a Convention offence]’. Resolution 1846 also appears to anticipate warships using this provision to deliver pirates to a port State. Nothing on the face of article 8 precludes the master of a ship other than the ship attacked using this provision. While article 2 of the SUA Convention states that it ‘does not apply … to a warship’, this provision was intended to prevent the Convention covering offences against military discipline. Neither the actual language used nor the intent behind it prevents this provision being applied by a warship. The receiving State may only refuse to accept delivery where it considers ‘that the Convention is not applicable to the acts giving rise to the delivery’.Footnote 64 Once a suspect is within a contracting port State's territory, that State is obliged to commence an investigation and either extradite the suspect or submit the case to national prosecutorial authorities.Footnote 65 To this end—unlike the law of piracy—State parties must have adequate national laws allowing prosecution of SUA Convention offences even in cases lacking an ordinary jurisdictional nexus with the prosecuting State.Footnote 66 The Convention's attractiveness as an alternative or supplementary basis for jurisdiction over Somali pirates is evident, especially given its coverage.Footnote 67 There is no evidence, however, that the article 8(1) mechanism has been used to ‘force’ a port State to receive suspects.
The appeal to the SUA Convention was reiterated in Resolution 1851, which in its preamble noted ‘with concern … the lack of capacity, domestic legislation, and clarity about how to dispose of pirates after their capture’ had led ‘to pirates being released without facing justice’. The preamble to Resolution 1897 similarly stresses ‘the need for States to criminalize piracy under their domestic law and to favourably consider the prosecution … of suspected pirates’, while operative paragraph 12 calls upon States ‘to ensure that all pirates handed over to judicial authorities are subject to a judicial process’. Both Resolutions further invite:
all States and regional organizations fighting piracy off the coast of Somalia to conclude special agreements … with countries willing to take custody of pirates in order to embark law enforcement officials (‘shipriders’) from the latter countries … to facilitate the investigation and prosecution of … [piracy suspects] …Footnote 68
Shiprider agreements were first developed in the maritime drug interdiction context.Footnote 69 A shiprider is a law-enforcement official of State A embarked on a vessel of State B who may, subject to specific treaty arrangements, authorize interventions aboard State A's vessels or in its waters, or arrest persons at sea under the law of State A. The last possibility is relevant here. If a piracy trial is to be conducted in, say, Kenya it could simplify matters if a Kenyan official made arrests and gathered evidence.
Operative paragraph 5 of Resolution 1851 also encourages consideration of a regional centre to ‘coordinate information relevant to piracy and armed robbery at sea’. This single information-sharing centre (ISC) model was adopted in the Regional Cooperation Agreement on Combating Piracy and Armed Robbery against Ships in Asia 2005 (ReCAAP).Footnote 70 The ReCAAP ISC can point to cases where information exchange has facilitated pirates' apprehension.Footnote 71 Its effectiveness, however, is easily over-stated: the decline in Malacca Strait piracy is largely attributable to increased, coordinated patrolling by littoral States.Footnote 72 Of greater use are single national points of contact available around-the-clock and capable of, for example, authorizing pursuits into territorial waters. A previous note described IMO-sponsored initiatives to promote such measures through a non-binding Memorandum of Understanding which envisaged authorities in Kenya, Tanzania and Yemen acting as reporting centres and regional States setting up single points of contact.Footnote 73 That MOU has now been translated, almost verbatim, into the Code of Conduct Concerning the Repression of Piracy and Armed Robbery against Ships in the Western Indian Ocean and the Gulf of Aden,Footnote 74 concluded at an IMO-sponsored meeting in Djibouti on 29 January 2009, and signed by 10 States (‘the Djibouti Code of Conduct’).Footnote 75 Operative paragraph 5 of Resolution 1851 also encourages support for regional capacity-building efforts assisted by the UN Office on Drugs and Crime (‘UNODC’) aimed at inter alia implementing shiprider agreements and the SUA Convention. A regional UNODC programme on piracy trials presently assists in ‘training prosecutors, locating and producing witnesses, facilitating international legal cooperation …, funding defence lawyers for pirates, and bringing prison conditions up to international standards.’Footnote 76
Importantly, all relevant Security Council Resolutions require counter-piracy action, wherever occurring, to be taken either consistently with ‘applicable’ international human rights lawFootnote 77 or ‘in conformity with international law’.Footnote 78 It is to such State practice this article now turns.
IV. COUNTER-PIRACY OPERATIONS AND STATE PRACTICE
A. The Internationally Recognized Transit Corridor
Various national and multinational counter-piracy missions now patrol the Gulf of Aden. The US-led Coalition of the Willing Combined Taskforce 150 (CTF-150), responsible for conducting maritime interdiction operations associated with Operation Enduring Freedom in Afghanistan,Footnote 79 commenced counter-piracy operations in August 2008. In January 2009 the Coalition established Combined Taskforce 151 (CTF-151), with an exclusive counter-piracy mandate.Footnote 80 Also in August 2008 Operation Enduring Freedom's combined forces established ‘a maritime security patrol area [‘MPSA’] in international waters off the Somali coast'.Footnote 81 The MPSA is a defined area within the Gulf of Aden, providing a common system of reference which allows naval forces in the Gulf to ‘de-conflict’ their activities.Footnote 82 Running through the MPSA is an internationally recognized transit corridor (the ‘IRTC’), established in August 2008 by the United Kingdom Maritime Trade Organization.Footnote 83 As of 1 February 2009, information for mariners using the IRTC is available through a secure website administered by the Maritime Security Centre (Horn of Africa), itself part of the EU counter-piracy mission Operation Atalanta. This is the EU's first joint naval operation, conducted by the EU Naval Force (EUNAVFOR), established in December 2008.Footnote 84 NATO's presence is conducted under Operation Ocean Shield: a counter-piracy mission to be carried out by the Standing NATO Maritime Groups.Footnote 85
The US, UK, Denmark and the EU have all concluded agreements for the transfer of captured pirates to Kenya for trial. NATO currently has no common legal framework to arrest pirates, nor any mechanism to transfer them to a third party for trial. If a NATO vessel seizes suspect pirates, the vessel ‘revert[s] to national control’ and their dispositionFootnote 86 becomes a question for the seizing warship's national State.Footnote 87
B. The Role of Human Rights in Detention, Transfer and Prosecution of Suspects
1. Threshold issues: applicable law, extra-territorial effect and attribution to international organisations
Captured piracy suspects must be detained, either aboard their own vessel or aboard a seizing warship while their disposition is determined. Where, as the US and France have done, a seizing State elects to try suspects before its own court no difficulty arises. Nor does any general rule of international law prohibit transfers of piracy suspects between equally competent jurisdictions. Thus Denmark sent suspects it captured after an attack on a Netherlands-Antilles registered vessel for trial in the Netherlands.Footnote 88 Formal extradition is not a requirement for such a transfer: consent may be granted by an exchange of diplomatic notes or other ad hoc agreement. As noted, the SUA Convention could be invoked to require port-State parties to receive delivery of captured suspects.
Though seizing warships in the Gulf of Aden generally offer to surrender piracy suspects to the national State of the attacked vessel or crew, in practice transfer to a regional State is the usual outcome where suspects are not simply released. Thus, over 100 suspects have been delivered to Kenya for trial, France routinely sends suspects for trial to Puntland,Footnote 89 China delivers suspects to the TFG and both the UK and EU have concluded transfer agreements with the Seychelles.Footnote 90 The detention, transfer and prosecution of suspect pirates all raise human rights law questions. Arguments have been made that Chapter VII authority to use ‘all necessary measures’ may implicitly authorize detention measures and displace or qualify conflicting treaty-based human rights obligations by virtue of article 103 of the UN Charter.Footnote 91 No such issue arises where, as here, relevant Resolutions expressly emphasise ‘applicable’ human rights law.Footnote 92
In dealing with suspected pirates States may thus be bound by obligations arising under applicable treaties, including the Convention against Torture (‘CAT’),Footnote 93 the Refugee Convention,Footnote 94 the European Convention on Human Rights (‘ECHR’)Footnote 95 and the International Covenant on Civil and Political Rights (‘ICCPR’),Footnote 96 not to return or expel persons to a place where they face a real risk of prohibited treatment (non-refoulement). A number of threshold questions arise. The first is the extraterritorial reach of these treaties. The position of the UN High Commissioner for Refugees is that the Refugee Convention ‘is subject to a geographic restriction only with regard to the country where a refugee may not be sent to, not the place where he or she is sent from’.Footnote 97 While it is sometimes said that the Convention can have no extra-territorial effect because it requires a protected person to have arrived in a Contracting State, this is incorrect: by definition a refugee has to be ‘outside’ their State of nationality, so one can only accurately say that someone who has not yet left their own state cannot be a refugee.Footnote 98 A person in a Convention State Party's custody at sea may thus not be returned to a place where they have a ‘well-founded’ fear of persecution on a relevant ground (race, religion, nationality, membership of a particular social group or political opinion).Footnote 99 This is likely to be of limited practical relevance: Somali pirates, motivated by money not politics, returned to Puntland may have little to fear from local governmentFootnote 100 and those that have committed a serious crime ‘outside the country of refuge’ are excluded from the Convention's protections.Footnote 101 Of more relevance is the CAT requirement that no party ‘expel, return (“refouler”) or extradite a person to another State where there are substantial grounds for believing that he would be in danger of being subjected to torture.’Footnote 102 ECHR article 3—as construed in the relevant case law—gives rise to a similar but broader requirement, prohibiting refoulement to face torture or ‘inhuman or degrading treatment or punishment’. The Committee Against Torture has found that parties ‘should apply the non-refoulement guarantee [in respect of torture] to all detainees in its custody’ including those extra-territorially detained;Footnote 103 and has applied the Convention in cases where a party exercises de facto or de jure control over persons aboard a vessel at sea.Footnote 104 Similarly, the European Court of Human Rights (‘the Strasbourg Court’) has applied the ECHR in high seas law enforcement cases.Footnote 105 Two tests for the extra-territorial application of the obligation to secure ECHR rights to those persons within a State party's ‘jurisdiction’ have been articulated: where a State exercises effective control over foreign territory; and where an individual comes under the authority and control of State agents acting extra-territorially.Footnote 106 The line between the two is not always well drawn in case law. Al-Saadoon and Mufdhi v United Kingdom concerned persons detained in Iraq. The Divisional Court found that ‘British forces had physical custody and control of the applicants and had it in their power to refuse to transfer them to the custody of [Iraqi authorities]’, even if to do so would breach international law.Footnote 107 The Strasbourg Court held that ‘given the total and exclusive de facto, and subsequently also de jure, control exercised by the United Kingdom authorities over the premises in question’ ECHR obligations were applicable.Footnote 108 Thus the Strasbourg Court appeared to hold that ‘total and exclusive de facto’ control over a place brings persons within a State's ‘jurisdiction’ for ECHR purposes. Whether this was considered a case of State agent ‘authority and control’ over an individual or ‘effective control’ over a (very restricted) space is not made clear. In the relevant cases it has become common simply to refer to the extra-territorial power exercised over embassies or aboard flagged ships and aircraft as clear cases where there is jurisdiction over persons therein.Footnote 109 As a result, it appears that the obligation to secure ECHR rights to those persons within a party's ‘jurisdiction’ will apply at the least to persons under that party's control in State-controlled spaces such ‘ships, embassies, and … detention facilities.’Footnote 110 Similarly, the International Covenant on Civil and Political Rights applies prima facie to the extraterritorial conduct of State organs, at least when exercising ‘effective control of an area outside … national territory’ or holding an individual in detention.Footnote 111 Thus, where a person is detained there appears potential for overlap between the ‘effective control’ and ‘authority and control’ tests.Footnote 112
What, then, will count as ‘effective control’ over a space or ‘authority and control’ over a person during maritime interdictions? Once persons are aboard a warship there could be no serious doubt they are within the flag State's jurisdiction. This holds true even if, for example, a person was rescued as a result of a safety of life at sea incident. However, a naval force may exercise other forms of ‘control’ over suspects. A warship or its small craft might first draw near a vessel and hail it (a so-called ‘friendly approach’); a boarding party might inspect and search a suspect vessel; where evidence of piracy is found, suspects aboard may be disarmed and/or prevented from sailing on; and suspects may be subsequently detained at sea either aboard their own craft or the warship. Considering these actions in turn, a ‘friendly approach’ is unlikely to involve ‘authority and control’. Visit and search is a finer question. UNCLOS article 110 allows only a limited investigatory function ‘which must be carried out with all possible consideration’. Provided this de jure authority is not exceeded, the degree of de facto control exercised is unlikely to engage the ECHR. However, if a State proceeds to exercise powers under UNCLOS article 105, it will be hard to argue that disarmed suspects effectively confined to a limited space (their vessel) are not within the seizing State's effective control. It would then appear irrelevant whether suspects are brought aboard a warship (as appears to be the Danish and Italian practice) or left in one of their own skiffs (the common UK practice). In the latter case, if the skiff has been disabled and cannot proceed under its own power, it must be under the warship's de facto control—even if no armed personnel are left aboard.
Multinational forces, however, raise further issues. The Strasbourg Court has been reluctant to find it has jurisdiction over Security Council Resolution-mandated peace-keeping operations or other multinational military actions.Footnote 113 The Court has been widely criticizedFootnote 114 for holding in Behrami and Saramati that UN Chapter VII authorization for an operation may mean that the actions of peacekeepers are attributable only to the UN and not to the peacekeeper's State of nationality,Footnote 115 at least when acting either as ‘subsidiary organs created under Chapter VII … [or] within the exercise of powers lawfully delegated by the Security Council’.Footnote 116 To this end the Strasbourg Court found a Chapter VII mandate vesting ‘ultimate authority and control’ in the Security Council determinative of the issue of attribution, despite the lack of any direct UN control over tactical operations (the ‘ultimate authority argument’).Footnote 117 This runs contrary to general principles of international law, where attribution follows not from the ultimate source of legal authority for an act but the exercise of effective control over it. The formulation adopted by the ILC in draft article 6 of its articles on the Responsibility of International Organizations is that:
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.Footnote 118
Despite quoting draft article 6 (at that time draft article 5) and its commentary, the Court misapplied the test. As the ILC Rapporteur on the subject has pointed out: ‘had the Court applied the criterion of effective control … it would have reached the different conclusion that the conduct of national contingents allocated to KFOR had to be attributed either to the sending [ie troop-contributing] State or to NATO.’Footnote 119 The ILC as a whole concluded, rather less bluntly, of Behrami ‘that when applying the criterion of effective control, “operational” control would seem more significant than “ultimate” control, since the latter hardly implies a role in the act in question.’Footnote 120
While it has been suggested that the test of ECHR ‘jurisdiction’ is autonomous from general international law rules of attribution/responsibility, it is hard to see what other meaningful standard exists for determining ‘effective control’.Footnote 121 The test must be a two-step one: first, is the individual within the ‘control’ of a prima facie State agent, such as a UK soldier; second, is it the UK—not some other entity—that ‘effectively controls’ that agent's act? Where both questions are answered ‘yes’, responsibility must vest in the agent's sending State. The first step involves the application of the ECHR test for jurisdiction and, for present purposes, may be considered to give rise to a rebuttable presumption that impugned conduct is attributable to the sending State. The acts of the UK soldier will be attributable to the UK unless it can be shown (under general principles of international responsibility) that the soldier was actually acting as the organ of another entity, such as an international organization. One cannot start with the legal authority of the international organization and work backwards: to do so is to take a purely theoretical approach at the expense of any analysis of the facts.Footnote 122 Legal authority is not the same thing as effective control; the latter is a fact-driven enquiry. The only thing that ultimate legal authority might suggest is the possibility of joint responsibility between a State and an international organization. One might be directly responsible for the wrongful conduct (ie where the official is acting as its organ), while the other might be in breach of a separate ‘due diligence’ or similar obligation to take positive steps designed to secure effective human rights protection.Footnote 123 Where, for example, an international organization is in a position to regulate acts in territory under its legal or effective control it might perhaps be held responsible for failure to take measures to prevent certain abuses.Footnote 124 This, however, does not exclude the responsibility of the troop-contributing State, especially in cases where it was the entity exercising ‘effective control’ over the impugned act. In sum, there is nothing to prevent the one wrongful act engaging the international responsibility of separate entities (albeit possibly in respect of distinct obligations).
Giving ECHR jurisdiction an entirely autonomous role simply allows a State to attempt to disclaim its effective control over individuals by claiming that ultimate legal authority for the acts complained of vests elsewhere. Thus the UK argued in Al-Saadoon that while its forces' actions in Iraq might at international law be attributable to the UK, it was only detaining and transferring prisoners at the behest of Iraqi courts or in exercise of Iraqi public authority.Footnote 125 The idea that the UK as a State was acting as Iraq's agent, or as bailee of Iraqi prisoners, is irrelevant to the question of factual control. Nonetheless, while it is counter-intuitive that a person aboard, say, a UK warship might not be within the UK's jurisdiction (and thus be outside ECHR protections), this may be precisely the result in certain multinational operations.
Any attribution-based analysis must commence from the proposition that neither the EU nor NATO are—presently—ECHR parties.Footnote 126 Where a unified command results in EU or NATO organs having ‘effective control’ over tactical-level conduct it may be that the ECHR does not apply, even if it would apply had the troop-contributing State been acting alone. Put another way, if a given warship is acting solely under the effective control of an international organization, a person in custody aboard cannot be within the flag State's effective control or ECHR jurisdiction. Thus even if a warship-contributing State could not consistently with its ECHR obligations transfer suspects to Kenya, its EUNAVFOR warships could perhaps do so where acting solely as EU organs. The relevant test of ‘effective control’ will be determined by asking which entity had ‘factual control’ over the ‘specific conduct’ in question.Footnote 127 While under EUNAFOR there is a unified EU command structure reaching down from the EU Political and Security Committee via an EU Operation Commander and then a Force Commander to theatre-level operations,Footnote 128 this alone does not tell us who determines the disposition of suspected pirates. Under the EU–Kenya Exchange of Letters, Kenya undertakes to ‘accept, upon the request of EUNAVFOR, the transfer of persons detained by EUNAVFOR in connection with piracy … for … investigation and prosecution.’Footnote 129 The question is whether this power is exercised by EUNAVFOR or national authorities.Footnote 130 If the decision is one for the EU Operation Commander alone, national responsibility might well not be engaged. If national authorities can make this decision, then suspect pirates are likely within the capturing State's ‘jurisdiction’. In practice, transfers to Kenya under the Exchange of Letters require the positive agreement of both the national authorities of the capturing warship and the EU Operation Commander. Under such conditions, a transfer decision cannot plausibly be considered an act of the EU alone. It is a situation of joint responsibility: both the national State of the capturing warship and the EU have it in their power to prevent any given transfer and both must concur for it to happen. As noted, in NATO operations a vessel seizing pirates reverts to national control and national authorities determine disposition.Footnote 131 If in practice disposition is determined by national authorities or requires their positive assent, then relevant suspects are within their ‘effective control’ or ‘authority and control’ and thus within their ECHR jurisdiction.
Even if a unified EUNAVFOR command did mean the ECHR was not directly applicable, several caveats should still be entered. First, there is no guarantee national courts will take this position regarding national human rights instruments. Whatever its flaws, the House of Lords decision in Al-Jedda distinguished Behrami and circumvented the Strasbourg Court's ultimate authority argument.Footnote 132 Second, while it would not give individuals the same procedural remedies as the ECHR, the EU would remain at the international level bound by general international law and many ECHR guarantees have customary international law equivalents, quite apart from their potential application as general principles of European law.Footnote 133 Third, there are national contingents in the Gulf of Aden that are not part of EUNAVFOR and are bound either by the ECHR (eg Russia) or by equivalent provisions in the International Covenant on Civil and Political Rights (eg Australia, Canada, India).
The relevant human rights issues to consider are thus: (1) legal authority to detain suspect pirates at sea and their right to be brought promptly before a judicial authority; (2) non-refoulement and transferring suspect pirates to regional States for prosecution; (3) the application of fair-trial rights in such transfers; and (4), in the ECHR context, the right to an effective remedy requiring the ability to challenge one's transfer.
2. Detention pending disposition and judicial supervision
Authority to detain piracy suspects at sea follows from powers of arrest. The general grant of authority to arrest persons aboard a pirate craft must encompass an incidental power to detain them.Footnote 134 A master's power to deliver a person suspected of a SUA Convention offence to another party's authorities must similarly include a power of interim detention.Footnote 135 The Security Council's calls for States operating on the high seas off Somalia to use ‘necessary means … for the repression of … piracy’ likely authorizes ‘necessary’ detention,Footnote 136 especially given the Council's express concern regarding ‘pirates being released without facing justice’.Footnote 137
A suspect pirate's right to be brought promptly before a judicial authority to rule on the legality of their detention, under ECHR article 5(3) or ICCPR article 9(4), generates complexities. Failure to consider in advance who will take jurisdiction over captured pirates may lead to problems, especially delay. In May 2009 it was reported that ‘[t]he Russian navy has been holding 29 suspected pirates aboard one of its warships for three weeks while it determines whether to prosecute them in Russia.’Footnote 138 In this respect, ECHR jurisprudence brings mixed blessings to the law of the sea. The leading authorities are Medvedyev Footnote 139 and Rigopolous.Footnote 140 In Medvedyev French authorities interdicted a Cambodian vessel suspected of drug smuggling on the basis of Cambodian consent; those on board were confined aboard their vessel while it was escorted to a French port, a voyage taking 13 days.Footnote 141 Rigopoulos similarly involved a Spanish high-seas interdiction of drug smugglers and a 16-day voyage to port. The Court held in both cases that such delays, where materially impossible to avoid, did not violate ECHR article 5(3).Footnote 142
However, Medvedyev introduces two complications through the Court's application of ECHR article 5(1): first, it held that there must be clear legal authority at national law to justify detention at sea, which was lacking in French law; and second, that there needed to be express textual authority at public international law to justify boarding the vessel, detaining persons aboard and subsequently prosecuting them.Footnote 143 The point to appreciate is that this is to apply the principle of legality twice: at both the national and international level. The Court found Cambodia's authorization of the ‘planned interception and all its consequences’, which was expressly granted ‘without restrictions or reservations’, was not sufficiently precise to justify detention.Footnote 144 A general public international lawyer might well be puzzled that this result could follow from an ostensibly plenary grant of authority.Footnote 145 The Court's reasoning is better exposed by its obiter comment that the general language of some drug interdiction treaties (granting authority to take ‘appropriate measures’) failed to protect individuals from arbitrary detention.Footnote 146 Indeed, the Court held of relevant provisions of both French law and international law that:
they do not regulate the conditions of deprivation of liberty on board ship, and in particular the possibility for the persons concerned to contact a lawyer or a family member. Nor do they place the detention under the supervision of a judicial authority …Footnote 147
While ECHR parties in their national implementation of law-enforcement treaties must obviously comply with their human rights obligations, it is erroneous to assume that such treaties (necessarily drafted for implementation in various legal systems) can realistically incorporate detailed human rights guarantees. To assume treaties cannot justify pre-trial detention without express words contemplating criminal penalties risks sweeping away enforcement powers under treaties ‘drafted without the precision we now expect from modern penal codes’.Footnote 148 While UNCLOS's provisions on piracy do contemplate criminal law sanctions (and thus should pass the ECHR's test),Footnote 149 the UN Narcotics Convention only refers to ‘taking appropriate action’ with flag State consent in cases of maritime drug smuggling.Footnote 150 Until Medvedyev, such ‘action’ had always been interpreted as encompassing arrest and prosecution where there was flag State consent. While the Strasbourg Court is right to insist on the principle of legality (nullem crimen sine lege) in national implementation of law enforcement treaties, to apply a principle of strict legality (nullem crimen sine lege stricta et scripta) to the treaties themselves is to needlessly undermine the enforcement provisions of other treaty regimes. It is erroneous to presume that the nullem crimen or nullem poena principle applies in the same manner at the international level as at the national level.Footnote 151 France had, at time of writing, appealed the case to the Grand Chamber.
3. A real risk of torture or mistreatment: diplomatic assurances and non-refoulement
States, including the UK, have attempted to facilitate transfer of individuals to States with questionable human rights records by seeking assurances that such persons will not be mistreated. The starting position under both the ECHR and the CAT appears congruent: mere words of assurance are insufficient. The question is whether assurances can provide ‘in their practical application [in individual cases], a sufficient guarantee that [a transferee] would be protected against the risk of [prohibited] treatment’.Footnote 152 This involves a wide-ranging contextual assessment, including any monitoring mechanisms established under assurances given.
There is perhaps a preliminary question in selecting the appropriate case law governing such transfers. ‘Transfer’ is sometimes considered a category apart: while deportation is a sovereign power exercised against non-nationals under domestic law and extradition ‘is a special institution combining rules of domestic law with the rules of international treaty or customary law’, ‘transfer’ may be seen as a residual category covering, inter alia, ‘forced movement[s] of individuals’ not covered by deportation or extradition.Footnote 153 Some EUNAVFOR participant States appear to be transferring suspects by simple executive action, while others have a process of judicial oversight. Should persons ‘transferred’ to Kenya from the sovereign jurisdiction of a warship be treated as analogous to aliens being deported/expelled from State territory or to persons being extradited? For ECHR purposes the question is irrelevant. Convention liability arises ‘by reason of … [a Contracting State's] having taken action which has as a direct consequence the exposure of an individual to the risk of proscribed ill-treatment [in a destination State or otherwise]’ irrespective of the legal mechanism by which that ‘exposure’ occurs. While one might be tempted to draw analogies with cases concerning the expulsion of persons from a British consulate or British forces ‘transferring’ detainees held abroad to local authorities they are distinguishable.Footnote 154 Such cases involve the potentially competing concurrent jurisdictions of the territorial ‘host’ State and the State acting extraterritorially. No such concurrency arises on the high seas.
Any a priori categorization may thus misstate the issue. The threshold question is whether a person is within an ECHR party's jurisdiction. Once that threshold is crossed, the case law of the Strasbourg Court makes it clear that, however categorised, any transfer of a person involving a risk of an ECHR violation will engage similar principles.
Regarding expulsion, the Strasbourg Court has stated in Chahal and Saadi that it is well-established that article 3 may be engaged ‘where substantial grounds have been shown for believing that the person in question, if expelled, would face a real risk of being [tortured or inhumanly or degradingly treated or punished]’.Footnote 155 In Chahal the UK sought to rely upon India's assurances that the applicant would not be ill-treated upon return. The Strasbourg Court found that despite reform efforts, and the Indian Government's evident good faith, torture in custody and human rights violations in Punjab remained ‘endemic’; therefore it was ‘not persuaded’ that the assurances given adequately guaranteed Mr Chahal's safety.Footnote 156 The later Saadi case confirms that mere words of assurance will not provide a sufficient guarantee of protection against mistreatment, at least in cases where there are credible reports of widespread torture and a failure to investigate such abuses.Footnote 157 The adequacy of assurances in their ‘practical application’ is an issue of fact for case-by-case determination.Footnote 158 Under what circumstances, then, might assurances sufficiently guarantee an individual's safety? The Committee against Torture suggests that when applying:
… its non-refoulement obligations under [CAT] … the State party should only rely on ‘diplomatic assurance’ in regard to States which do not systematically violate the Convention's provisions, and after a thorough examination of the merits of each individual case. The State party should establish and implement clear procedures for obtaining such assurances, with adequate judicial mechanisms for review, and effective post-return monitoring arrangements.Footnote 159
On this approach an acceptable assurances regime should include: individual merits review, a clear procedure for obtaining assurances, opportunity for judicial review and ‘effective post-return monitoring’. Each of these measures will be a strong indicia of assurances' effectiveness, though they may not all be either necessary individually or sufficient cumulatively in a given case. The suggestion that assurances should not be used regarding States which torture ‘systematically’ is discussed below.
Current UK practice emphasizes post-return monitoring. Assurances sought in deportation cases now include:Footnote 160
[w]here possible, … independent monitoring arrangements providing for a trained human rights monitor to accompany the individual back from the UK to their country of return, to provide contact details and be contactable 24/7, to make contact weekly with the individual or their next of kin for the first year after the person returns, to have frequent, without notice, private visits to the person if in detention, and to arrange for prompt medical inspections at any time if they have concerns.
This is coupled with detailed investigation of post-arrival procedures: ‘where they would go, … be detained, … in which Court they would be tried … and where they would be held if convicted’.Footnote 161 While the words ‘where possible’ leave room for concern,Footnote 162 if diligently executed alongside ordinary administrative procedures and access to judicial review, such arrangements could satisfy the Committee against Torture's strictures and meet the Chahal test. Further, UK practice is to take steps to place assurances ‘at the heart of the bilateral relationship’, reinforcing that their breach will have serious diplomatic consequences.Footnote 163 Evidence of such communications and their effect will be relevant in assessing the value of assurances given.Footnote 164
The question remains whether recourse to assurances is a priori impermissible regarding States which torture ‘systematically’, the argument being that there are regimes with such a poor human rights record that assurances should never be accepted as curing the obvious risks a transferred person would face. Similarly, the Strasbourg Court in Saadi indicated that even had Tunisia given the assurances sought this might still have been insufficient.Footnote 165 Clearly there will be cases where it would be hard for any system of assurances to preclude real risks of torture. This, however, simply describes the outcome of a legal evaluation, not an a priori standard. Certainly, CAT requires serious scrutiny of any proposed transfer to a State where there are ‘consistent pattern of gross, flagrant or mass violations of human rights’.Footnote 166 However, such violations do not make deportation absolutely impermissible, they are simply ‘relevant considerations’ for competent authorities in determining whether ‘substantial grounds [exist] for believing’ that an individual transferee risks torture.Footnote 167 That said, one would obviously expect very strong justifications if such a deportation were ever to succeed. Whether assurances can be effective is a question of fact requiring case-by-case assessment of ‘the circumstances obtaining at the material time’Footnote 168 including ‘the human rights record of the state, the personal risk faced by the claimant, any assurances [given] … and their worth’ especially given the destination State's ‘ability … to control its own security forces’.Footnote 169 In sum, the idea that deportation with assurances should not be used regarding States which systematically torture is useful as a normative standard by which to judge outcomes or as a ‘red flag’ to decision makers that heightened scrutiny will be applied. It is not, however, a preliminary threshold that courts have applied to preclude transfers.Footnote 170
How might such considerations apply regarding transfers between a warship and a state such as Kenya? Formal assurances and post-transfer monitoring are the easy part. The preamble to the EU-Kenya Exchange of LettersFootnote 171 refers expressly to the ICCPR and the Convention Against Torture, and the parties undertake to:
… treat persons transferred …, both prior to and following transfer, humanely and in accordance with international human rights obligations, including the prohibition against torture and cruel, inhumane and degrading treatment or punishment, the prohibition of arbitrary detention and in accordance with the requirement to have a fair trial.Footnote 172
The first post-transfer monitoring mechanism established is documentary:
Kenya will … keep an accurate account of all transferred persons, including, but not limited to, … the person's physical condition, … their places of detention, any charges against him and any significant decisions taken in the course of his prosecution and trial.Footnote 173
These records must be made available to the EU or EUNAVFOR upon written request.Footnote 174 Other provisions require that:
• Kenya notify EUNAVFOR of the transferred person's place of detention, ‘any deterioration of his physical condition and of any allegations of alleged improper treatment’;
• EU and EUNAVFOR representatives ‘will have access’ to transferred persons while in custody and ‘will be entitled to question them’;
• ‘[n]ational and international humanitarian agencies will, at their request, be allowed to visit [transferred] persons’;Footnote 175 and
• the death penalty is expressly prohibited.Footnote 176
The Exchange of Letters provides for further implementing arrangements to be concluded.Footnote 177 Preferably, these should provide for something closer to the UK's preferred mechanism of third-party weekly visits, ‘frequent, without notice, private visits’, and prompt medical inspections. Presently pirate trials in Kenya are usually monitored—at least in the court room—by representatives of both EUNAVFOR and the seizing warship's flag State.Footnote 178
Even if such measures were in place, questions would remain concerning judicial supervision of transfers and human rights in Kenya's prison system. As noted, Italy has brought captured pirates before a judge by video conference for disposition. On 25 May 2009, an Italian Giudice d'Indagine Preliminare (pre-trial or investigating judge) ordered the transfer of nine pirates captured by the Italian warship Maestrale to an Italian jail.Footnote 179 This is significant as Meastrale is part of EUNAVFOR and the Exchange of Letters came into provisional application on 6 March 2009. As noted above, national authorities have a significant degree of control over transfers to Kenya, as such transfers require the consent of both the seizing State and the EU Operation Commander. What the Italian practice illustrates is that decisions regarding suspects' disposition or transfer may occur under a variety of national legal procedures, ranging from no independent review to close judicial supervision. This could of itself be cause for concern. The second question is whether torture is so widespread in Kenya's prisons that assurances are unlikely to effectively guarantee a transferee's safety. The Committee Against Torture in 2008 noted:Footnote 180
… with deep concern the numerous and consistent allegations of widespread use of torture and ill-treatment of suspects in police custody [in Kenya] … [and] the challenges reported by … [Kenya] in providing people under arrest with the appropriate legal safeguards, including the right to access a lawyer, an independent medical examination and the right to contact family members.
The Committee Against Torture also noted that the Kenya National Commission on Human Rights faced difficulties in ‘freely access[ing] and monitor[ing] places of detention’ and ‘the dire conditions … in Kenyan prisons’ as regards ‘overcrowding, … health services and high levels of violence’.Footnote 181 While Kenya has committed itself to wide-scale justice and penal system reform, the African Peer Review Mechanism has noted Kenya's limited resources for achieving those goals.Footnote 182
All of this could be fatal to the Strasbourg Court or Committee Against Torture upholding present transfer arrangements. Under similar conditions the Strasbourg Court has held assurances without monitoring were an insufficient ‘guarantee against the serious risk of ill-treatment’ where an extradited person would be subjected to very poor prison conditions in a State where there were credible reports of police torture.Footnote 183 This could be detrimental to prison and justice reform in Kenya since Kenya's most effective avenue for securing rule-of-law aid at present is its willingness to prosecute pirates. Given that there are presently 101 piracy suspects in Kenyan prisons and 10 have now been convicted and sentenced,Footnote 184 by the time of any ECHR challenge the Strasbourg Court should be able to assess how transfers are working in practice. If it is demonstrable that piracy suspects are being humanely detained and treated and tried fairly, then the transfer mechanism could be found compatible with non-refoulement obligations, provided there is some mechanism for independent review of transfers and preferably for post-transfer monitoring. It certainly behoves States transferring pirates to Kenya for trial to provide material support and help improve prison conditions. Some assistance, including prison reform, is already occurringFootnote 185 and preventing mistreatment of these highly-publicised prisoners is in all the parties' interests.
4. Fair trial guarantees
The Exchange of Letters incorporates express guarantees drawn from ICCPR article 14, also found in ECHR Articles 5 and 6, requiring:
• that a transferred person be brought promptly before a judicial authority to determine the legality of their detention;Footnote 186
• a right to ‘trial within a reasonable time or to release’;Footnote 187
• a right to trial in ‘a fair and public hearing by a competent, independent and impartial tribunal established by law’;Footnote 188
• the presumption of innocence;Footnote 189
• the right to be informed of the charges against one, to have adequate opportunity to prepare a defence, to be tried in one's presence (with legal assistance where necessary), to examine the evidence, to have an interpreter and a right to silence;Footnote 190 and
• a right of review as to sentence by a higher Kenyan court.Footnote 191
By what standard should these assurances be judged? While the article 3 non-refoulement cases seemingly give quite generous rights protection to transferees, UK courts have held that ‘a high threshold test’ applies in deportation cases concerning other ECHR rights.Footnote 192 This follows from the fact that
the [Strasbourg] Court has not to date found that the expulsion or extradition of an individual … [would violate] Article 6 of the Convention, [although] it has on frequent occasions held that such a possibility cannot be excluded where the person being expelled has suffered or risks suffering a flagrant denial of a fair trial …Footnote 193
Under this test foreign systems of justice may fall quite short of ECHR standards without creating a ‘flagrant denial’. Indeed, much turns on defining ‘flagrant denial’. The only substantive elaboration of the concept is found in the joint partly dissenting opinion of Judges Bratza, Bonello and Hedigan in Mamatkulov and Askarov, who would have found a breach of article 6 in that case.Footnote 194 In their view ‘the word “flagrant” is intended to convey … a breach of the principles of fair trial … so fundamental as to amount to a nullification, or destruction of the very essence, of the right guaranteed’.Footnote 195 The House of Lords has thus held that to engage fair trial rights in a deportation case, ‘[i]t will be necessary to establish at least a real risk of a flagrant violation of the very essence of the right’.Footnote 196 The minority in Mamatkulov would have found this standard met where there were ‘credible grounds for believing that self-incriminating evidence extracted by torture was routinely used to secure guilty verdicts and that suspects were very frequently denied access to a lawyer’.Footnote 197 These violations were compounded, in their view, by the failure to seek or obtain assurances ‘safeguard[ing] the applicant's continued contact with their legal representatives or … [their] access to … independent lawyers of their own choosing’.Footnote 198 However, the real risk of such violations was insufficient for the majority in Mamatkulov; and the House of Lords in applying even the stricter minority test found no article 6 violation in a case involving ‘a very real risk’ that evidence obtained by torture would be used against a person, as it could not be concluded on the balance of probabilities that such evidence would be used.Footnote 199 Such high thresholds have led Elliott to conclude that ‘the flagrant denial of justice test is largely if not wholly illusory’ in extradition cases.Footnote 200
It seems unlikely that pirate trials in Kenya would violate article 6. Apart from the high threshold imposed by ECHR jurisprudence, there are few obvious grounds to impeach their fairness. Despite the general backlog in Kenya's judicial systemFootnote 201 they commence promptly (in as little as six weeks), the pirates are represented by local lawyers (often with foreign legal assistance),Footnote 202 translators have been paid for by capturing States, EU-provided assistance has included ‘computers and money to bring more qualified Kenyan judges to Mombasa’ and diplomatic observers are routinely present.Footnote 203 Further, the Kenyan justice system requires oral testimony, and great lengths have been gone to in order to secure the presence of witnesses such as an attacked ship's master.Footnote 204
5. Right to an effective remedy
The final question is whether transferring suspects to a State where there is a real risk of prohibited treatment without some review process violates the right to an effective remedy. As the Strasbourg Court has put it:Footnote 205
[G]iven the irreversible nature of the harm that might occur if the risk of ill-treatment materialised …, the notion of an effective remedy under Article 13 requires independent scrutiny of the claim that there exist substantial grounds for fearing a real risk of treatment contrary to Article 3. … Such scrutiny need not be provided by a judicial authority but, if it is not, the powers and guarantees which it affords are relevant in determining whether the remedy … is effective.
Thus, there ‘must be a remedy which may prevent’ transfer and the relevant national authority must be able to ‘examine the Convention compatibility of the … [transfer] order prior to its implementation.’Footnote 206 A key requirement is that any application for review must either have an automatic ‘suspensive’ effect or allow time for a ‘stay of execution’ to be sought.Footnote 207 While it is clear that the mechanism need not be judicial, where a judicial remedy (with suspensive effect) is available it will usually meet the article 13 standard. The judicial oversight mechanism established by Italy would thus appear sufficient. Similarly, the Netherlands apparently considers Dutch extradition law applicable and has law-enforcement personnel and ‘an assistant-district attorney’ embarked on government vessels in the region.Footnote 208 The UK, though, has no specific oversight mechanism in place regarding transfers of piracy suspects. However, as Al-Saadoon illustrates, foreign detainees held extraterritorially by UK forces may challenge their proposed transfer through judicial review proceedings and seek an interim injunction to prevent their transfer pending hearing.Footnote 209 This too should be sufficient to meet the requirements of article 13.
V. THE WAY FORWARD
So long as national authorities determine the disposition of seized piracy suspects, they will be bound by applicable human rights law. States cannot escape such obligations through acting as part of EU or NATO forces, unless the relevant international organization effectively controls suspects' disposition. Even then, a State may nonetheless have a separate obligation to take positive steps to ensure that the mission to which it is contributing troops will be conducted in a manner complying with relevant human rights standards as discussed above. ECHR law has been given the greatest space here, as it holds out the greatest prospect for enforcement. The potentially applicable ECHR rights include the right to be brought promptly before a judge, fair trial rights, the prohibition on refoulement to face torture or other impermissible treatment, and the subsidiary right to an ‘effective remedy’. The first two are most easily dispensed with. Inevitable delay attendant on bringing a suspect to port is unlikely per se to breach ECHR obligations, but States should consider how to avoid unnecessary delays ahead of time. As noted, the standard for review in deportation/extradition cases alleging a risk of fair-trial rights violations is extremely (perhaps impossibly) high, and there is little to suggest pirate trials in Kenya are being unfairly conducted.
Obligations of non-refoulement under the CAT or ECHR raise serious concerns in transferring pirates to any regional State, including Somalia, to face trial. Prison and justice-system conditions in Kenya may be comparable to those in ECHR cases where the Strasbourg Court has rejected deportation based on assurances alone. For a system of transfers to be able to withstand ECHR review, State parties will have to be able to establish—where individuals transferees can demonstrate a risk of mistreatment—the absence of that risk in that individual case.Footnote 210 As a practical matter, this could best be achieved by a system applicable in all cases aimed at minimizing such risks. Obvious possible measures include aid to improve conditions in regional prisons and a rigorous monitoring programme throughout pre- and post-trial detention. The EU-Kenya Exchange of Letters includes a series of rights of access to prisoners capable of being used for the latter but without any indication that they will be so used in all cases.Footnote 211 While the lack of an automatic system of independent prison visits is concerning, it may not be fatal so long as effective monitoring occurs in practice. The more important hurdle to overcome is the right to an effective remedy: there must be some facility for scrutinising any claims by suspects regarding potential mistreatment in the destination State. It is not clear that all States within EUNAVFOR presently making transfers to Kenya have applicable independent review mechanisms in place, although some clearly do.
The more it can be demonstrated that transfers meet both universal and European human rights requirements, the more objections to them will fall away. One suggestion to this end has been the establishment of an international or regional piracy court.Footnote 212 The problems with such an approach are legion and obvious. The expense of ad hoc tribunals is something of which the international community has become justifiably wary. While piracy trials should be significantly less complex than prosecutions for war crimes, crimes against humanity and genocide any international tribunal will remain significantly more expensive than trials in a national court. The mere fact of less complex trials is also no guarantee that early trials will not be significantly delayed by appeals on preliminary points, as the international tribunal fleshes out its own procedural law and elements of crimes. It is a mistake to believe these can be exhaustively settled in advance, as proceedings before the International Criminal Court demonstrate. An international court may also not serve to strengthen local justice systems. When it inevitably closes down its accumulated experience will be dispersed as international staff return to foreign capitals, rather than being folded back into local judicial systems. In sum, in this age of international ‘tribunal fatigue’ such a proposal seems unlikely to attract wide support.Footnote 213
A preferable model would be substantial international assistance for trials in the region. This could take the form of an ‘international’ or ‘extraordinary’ chamber within the courts of a regional State.Footnote 214 The advantages of a national court applying national law under established rules of evidence and procedure over inventing everything from scratch are manifest. As discussed above, shiprider arrangements could also facilitate the use of local justice systems. Support should also be provided to strengthen (or create) regional coast guards or a regional counter-piracy maritime force. Hearteningly, a regional training centre is to be established in Djibouti and Japan has already approved US $15 million in aid towards implementing the Djibouti Code of Conduct.Footnote 215
