Yesterday the High Court of England and Wales, per Mr Justice Leggatt, delivered a comprehensive judgment in Serdar Mohammed v. Ministry of Defence  EWHC 1369 (QB), holding that the United Kingdom lacks detention authority under international humanitarian law/law of armed conflict with regard to individuals it captures in the course of the non-international armed conflict in Afghanistan, and that any detention of such individuals longer than 96 hours violates Article 5 ECHR, as well as relevant Afghan law. The judgment is on any account a heroic effort, with the single judge grappling with a host of complex, intertwined issues of international law and acquitting himself admirably in the process. Para. 6 contains a summary of the judgment for those who don’t want to read the whole thing.
Here are some of the highlights of the Court’s analysis:
(1) The ECHR applies extraterritorially to any person detained by the UK in Afghanistan.
(2) Derogations under Article 15 ECHR could also be used in an extraterritorial context.
(3) The detention of SM by UK forces in Afghanistan was attributable to the United Kingdom, and not to the UN .
(4) No conflict arose between relevant UNSC resolutions, which did not authorize SM’s continued detention, and Article 5 ECHR, and Article 103 of the Charter was inapplicable.
(5) SM’s detention was not authorized by IHL either, since IHL in NIACs contains no detention authority, and cannot prevail over Article 5 ECHR as lex specialis.
(6) SM’s detention violated Article 5 ECHR. While the detention up to 96 hours was Article 5-compliant, the 110 days that SM spent in UK detention were not.
The Court makes it clear that the position the UK government found itself in is largely its own doing (para. 417 ff). This is exactly right. The government’s own legal advisers informed it of the limited extant legal authority for prolonged detention. The UK government failed to enact its own domestic legislation on detention in Afghanistan, or to come to different arrangements with Afghan authorities. Similarly, the UK government chose not to derogate from the Convention, preferring instead to argue that the Convention does not apply. And now that this strategy has failed (and on several levels), much of what it has been doing is exposed as unlawful.
I imagine that the judgment will be appealed, and we shall we see what happens there. But whatever the appellate courts’ conclusions, I can only hope that their judges will show as much diligence and analytical precision as Mr Justice Leggatt.
Here are the highlights, with some commentary:
(1) The ECHR applies extraterritorially to any person detained by the UK in Afghanistan, under the ECtHR Al-Skeini judgment and the personal model of Article 1 ECHR jurisdiction as authority and control over individuals (para. 166 ff). The judge thought of Al-Skeini as follows:
136. A disappointing feature of the judgment of the European Court in the Al-Skeini case is its lack of transparency in dealing with its previous decision in the Bankovic case. Nowhere did the Court confront or expressly acknowledge the fact that it was departing from its previous approach or explain why it was doing so. The Bankovic case is not even mentioned except for citations to it in some footnotes.
137. It is clear, however, that in the Al-Skeini case the European Court has indeed departed from its approach in the Bankovic case on all the five points which I mentioned above. In particular:
i) The Court has now endorsed a principle of jurisdiction based on the exercise of effective control by a state over an individual;
ii) The Court has expressly resiled from the notion that Convention rights constitute a single, indivisible package and has said that they can be “divided and tailored”;
iii) The Court held that jurisdiction under article 1 is not limited to the territory of states which are parties to the Convention;
iv) In endorsing an approach which goes well beyond what the Court had found in the Bankovic case to be ordinary meaning and original intention of Article 1, the Court has effectively treated Article 1 as a “living instrument”;
v) Although the Court continued to pay lip-serve to the notion that jurisdiction is “essentially territorial” and that extraterritorial jurisdiction is exceptional, it is difficult to see how this can remain so when jurisdiction arises wherever in the world a state exercises effective control over an individual.
The judge did not put much stock in the ‘public powers’ concept that the ECtHR used to prop up the personal model of jurisdiction and prevent its collapse (see more here), but he was fully aware that it is prone to collapsing into the proposition that the state has the duty to respect human rights whenever it has the factual ability to violate them:
141. The decision of the European Court in the Al-Skeini case leaves many unanswered questions which will no doubt have to be worked out in later cases. For example, it is unclear whether, once jurisdiction is understood to rest on the exercise of control over individuals, there is any stopping point short of what the European Court in the Bankovic case saw as the logical conclusion that jurisdiction under Article 1 exists whenever an act attributable to a contracting state has an adverse effect on anyone anywhere in the world; and if so, what that stopping point is. In the present case, however, such difficult questions do not arise because the facts fall squarely within one of the core examples of the control principle set out in the Al-Skeini case and not merely within its penumbra.
The judge thus concluded that the claimant was within the UK’s jurisdiction, since detention undoubtedly qualified as an exercise of physical power and control over him (paras. 147-148). Note that the UK government’s immediate reaction to Al-Skeini was to say that it was contained to the unique facts of Iraq, and the MoD lawyers similarly tried to distinguish Al-Skeini in several different ways, all of which the Court ultimately found unpersuasive. The government’s strategy to deny the ECHR’s applicability to Afghanistan thus seems likely to fail, not just in the further proceedings in this case but in others as well.
(2) Disagreeing with the dicta of the House of Lords in Al-Skeini and the UK Supreme Court in Smith, the judge considers that derogations under Article 15 ECHR could also be used in an extraterritorial context (paras. 153-157), finding that:
155. Article 15 accordingly permits a state, within defined limits, to derogate from its obligations under the Convention “in time of war or other public emergency threatening the life of the nation.” This wording, however, (in particular the word “other”) tends to suggest that Article 15 was not intended to apply to a war overseas which does not threaten the life of the nation. That is no doubt because those who drafted the Convention did not envisage that a state’s jurisdiction under Article 1 would extend to acts done outside its territory. Now that the Convention has been interpreted, however, as having such extraterritorial effect, it seems to me that Article 15 must be interpreted in a way which reflects this. It cannot be right to interpret jurisdiction under Article 1 as encompassing the exercise of power and control by a state on the territory of another state, as the European Court did in the Al-Skeini case, unless at the same time Article 15 is interpreted in a way which is consonant with that position and permits derogation to the extent that it is strictly required by the exigencies of the situation.
I think this is perfectly sensible. Recall also the pending ECtHR Hassan case, where the Court may pronounce itself on extraterritorial derogations. I would only part ways with the judge when he says that the drafters of the Convention did not envisage extraterritorial Article 1 jurisdiction, since we really have absolutely no idea what the drafters of the Convention wanted or intended.
(3) The detention of SM by UK forces in Afghanistan was attributable to the United Kingdom, and not to the UN (para. 158 ff). Here we have an extended discussion of the Behrami and Saramati and Al-Jedda attribution saga. While the judge thought that the distinguishing in Al-Jedda between the situations in Iraq and Kosovo was not wholly persuasive, and while he did not reject Behrami outright (which I think he could and should have done), he considered that:
178. In these circumstances, although I do not find the question easy, I consider that the UN Security Council has “effective control” (and “ultimate authority and control”) over ISAF in the sense required to enable conduct of ISAF to be attributed to the UN. Thus, if the detention of SM had been authorised by COMISAF (in the way that COMKFOR authorised the detention of Mr Saramati) and a claim had been brought against the state from whose armed forces COMISAF was drawn on the basis that that state was in breach of Article 5 of the Convention, I would expect the European Court to hold that the detention was not attributable to the respondent state, applying the same analysis as it did in the Behrami and Saramati cases. (I am assuming for the purpose of this hypothetical case that COMISAF at the relevant time was an officer in the armed forces of a state which is a contracting party to the Convention.)
However, that notwithstanding, SM’s detention was still attributable to the UK because it was ordered not by the ISAF commander, but by UK commanders and ministers (para. 185), hence:
187. In these circumstances, it is in my view quite clear that the detention of SM is attributable to the United Kingdom. It is unnecessary for me to consider the possibility of joint responsibility, as I think it equally clear that the acts involved in the detention of SM are not attributable to ISAF or the UN.
(4) Accepting that he was bound by the House of Lords’ Al-Jedda decision that UN Security Council resolutions can prevail over the ECHR by virtue of Article 103 of the Charter (paras. 209-211), and that the UNSC resolutions relevant to Afghanistan authorized ISAF to ‘take all necessary measures,’ which include lethal force in self-defence, the judge considered that these measures must also include capturing adversaries on the battlefield (para. 218). However:
219. I accept this argument so far as it goes. In particular, I accept that the UNSCRs relating to Afghanistan were plainly intended to authorise the use of lethal force at least for the purposes of self-defence. I also accept that in these circumstances it must be the case that ISAF personnel were authorised to take the lesser step of accepting the surrender of individuals who were believed to pose an imminent threat to them or to the civilian population. I see no necessary implication, however, that this authorisation was intended to give ISAF a power to continue to hold individuals in detention outside the Afghan criminal justice system after they had been arrested and therefore ceased to be an imminent threat.
220. As mentioned earlier, the mandate of ISAF was to assist the Afghan government in the maintenance of security. In addition, the UNSCRs expressly affirmed the sovereignty, independence and territorial integrity of Afghanistan and recognised that the responsibility for providing security and law and order throughout Afghanistan resided with the Afghan authorities. In these circumstances, and in circumstances where (as discussed in part IV of this judgment) ISAF had no power under Afghan law to detain individuals other than to hand them over immediately to the police or a prosecutor, I can see no reason to interpret the authorisation to “take all necessary measures” to fulfil the ISAF mandate as permitting detention for any longer than was necessary to deliver them to the Afghan authorities.
221. Nor can I see any reason to interpret that authorisation as permitting detention by ISAF which violated international human rights law. In ascertaining the scope of the relevant authority, it seems to me that I must take into account the principles endorsed by the European Court in Al-Jedda v United Kingdom (2011) 53 EHRR 23. Section 2(1) of the Human Rights Act requires me to do so in circumstances where the opinion of the European Court is relevant to the question that I have to determine concerning the scope of the claimant’s Convention rights. As mentioned, the European Court considered there to be a presumption that, unless it uses clear and unambiguous language to the contrary, the Security Council does not intend states to take measures which could conflict with their obligations under international human rights law. In the Al-Jedda case the European Court did not regard even the language used in UNSCR 1546 and the letter from Mr Powell annexed to it which expressly referred to internment as sufficiently clear and unambiguous to override this presumption. In the resolution applicable in the present case there is no express reference at all to internment or detention. Although I consider that a power to detain is implied, there is nothing in the language of UNSCR 1890 which demonstrates – let alone in clear and unambiguous terms – an intention to require or authorise detention contrary to international human rights law.
Accordingly, the judge concluded that detention up to 96 hours, per the ISAF detention policy, for the purpose of transferring the detainees to Afghan custody, was included within the mandate, but that longer detention for the purpose of intelligence-gathering, under the UK’s own detention policy, was not (paras. 224-227). No conflict thus arose between UNSC resolutions and Article 5 ECHR, and Article 103 of the Charter was inapplicable.
(5) SM’s detention was not authorized by IHL either, since IHL in NIACs contains no detention authority, and cannot prevail over Article 5 ECHR as lex specialis (para. 228 ff). The judge noted that:
239. Neither CA3 nor Article 5 of AP2 contains any express statement that it is lawful to deprive persons of their liberty in an armed conflict to which these provisions apply. All that they do is to set out certain minimum standards of treatment which must be afforded to persons who are detained during such an armed conflict. The MOD argues, however, that a power to detain is implicit in CA3 and AP2.
240. This argument has the support of some academic writers and of the International Committee of the Red Cross (“ICRC”). Thus, Jelena Pejic, the legal advisor to the ICRC, has written: “Internment is … clearly a measure that can be taken in non-international armed conflict, as evidenced by the language of [AP2], which mentions internment in Articles 5 and 6 respectively …”
The judge then proceeds to disagree with the implicit NIAC detention authority argument for five reasons (note, however, that I think that the paper by Jelena Pejic was written in her personal capacity). First, because powers to deprive individuals of liberty should not be inferred indirectly or by implication (para. 242). Second, because the fact of detention in NIAC is not the same as authorization of such detention (para. 243). Third, the relevant treaty provisions, such as CA3, are of humanitarian character only and refer to treatment in detention, not detention itself (para. 244). Fourth, IHL-based authority to detain in NIACs would mean that non-state actors, i.e. rebels and other organized armed groups, would also have authority to detain, which would be anathema to most states (para. 245). Fifth, neither CA3 nor AP2 specify who can be detained, when and for how long.
I think the judge got it exactly right on this point. Detention authority in NIACs exists only on the basis of domestic law, and here there was no such basis in either Afghan law or UK law.
He then proceeds to look at the lex specialis principle, and correctly identifies its three different variants (total displacement, partial displacement, and as a tool of interpretation – para. 269 ff). He finds the total displacement version (i.e. that IHL applies to the complete exclusion of human rights) untenable, and considers that:
279. The difficulty is increased by the fact that the Convention contemplates and makes provision within itself for situations of war. Thus Article 15 (quoted at paragraph 154 above) permits a state to derogate from its obligations under the Convention in a time of emergency. The clear and necessary implication of Article 15 is that the Convention continues to apply in a situation of armed conflict except to the extent that (a) a contracting state derogates from its obligations under the Convention and (b) such derogation is permitted by Article 15.
(Recall again his previous acceptance of extraterritorial derogations). He similarly finds the partial displacement conception not entirely persuasive:
284. At least arguably, however, even in a case where such a conflict of obligations occurs, the only way in which the European Court or a national court required to apply Convention rights can hold that IHL prevails over Article 5 is by applying the provisions for derogation contained in the Convention itself, and not by invoking the principle of lex specialis. In considering the extent to which derogation is “strictly required by the exigencies of the situation” and therefore permissible, Article 15(1) expressly allows regard to be had to a state’s “other obligations under international law”, which plainly includes IHL. The obligation of a state to comply with IHL would thus be a compelling justification for derogating from Article 5 in relation to the detention of POWs during an international armed conflict. However, in circumstances where the Convention itself defines the conditions in which and the extent to which derogation from its obligations is permitted, and makes specific provision for derogation in time of war, it is difficult to see that there is any room for the lex specialis principle to operate as a basis for disapplying the Convention when it conflicts with IHL.
However, the judge notes that he does not have to pronounce on the validity of the partial displacement variant of lex specialis in the absence of any derogation, since in his view IHL in NIACs does not provide any express detention authority anyway (para. 287). He proceeds to find the more modest, interpretative version of lex specialis (i.e. that IHL is to be taken into account in interpreting the relevant provisions of human rights treaties) to be entirely unobjectionable, while noting the difference between the open arbitrariness standards used in Article 9 ICCPR and the closed list of grounds for detention in Article 5 ECHR:
291. Unlike Article 9(1) of the ICCPR, however, Article 5(1) of the Convention is much more specific and prohibits arrest or detention “save in the following cases” which are then exhaustively defined. Given the specificity of Article 5, there is little scope for lex specialis to operate as a principle of interpretation. Furthermore, in view of my conclusion that in a non-international armed conflict IHL does not specify grounds for detention or procedures to be followed, there are in my view no relevant rules of IHL with which to try to harmonise the interpretation of Article 5.
292. As I see it, where IHL would be relevant in applying the Convention is in situation where a state resorted to measures derogating from Article 5 on the basis that it was involved in a non-international armed conflict. In such circumstances it could be argued that, under customary IHL, certain fundamental guarantees against the arbitrary deprivation of liberty should still be respected (see paragraph 260 above). To the extent that such guarantees form part of customary IHL, derogations from them would be “inconsistent with [a state’s] other obligations under international law” and therefore not permitted by Article 15.
On this, again, the judge is in my view entirely correct.
(6) SM’s detention violated Article 5 ECHR. While the detention up to 96 hours was Article 5-compliant, the 110 days that SM spent in UK detention were not. Purely preventive security detention violated Article 5(1) ECHR absent a derogation (para. 310). SM was not brought promptly before a judicial officer, not given the possibility of judicial review of his detention. In particular:
343. I referred in part VI above to the principle endorsed by the European Court in the Al-Skeini case that the obligation under Article 1 to secure Convention rights can be “divided and tailored” to reflect the extent of the relevant control on which extraterritorial jurisdiction is founded. Counsel for the MOD submitted that, in accordance with this principle, if Article 5(3) cannot be interpreted sufficiently flexibly to accommodate the fact that the UK had no power to bring SM before an Afghan court, then it must be regarded as inapplicable.
344. As already mentioned, the MOD has not shown that it was impracticable either to arrange for detainees awaiting transfer to an Afghan prison to be brought before an Afghan judicial officer in liaison with the Afghan authorities or, alternatively, to arrange for detention reviews to be carried out by a British judicial officer. Even if, however, neither of these arrangements was feasible, there is in my view sufficient flexibility in Article 5(3) by reason of the fact that it did not require the UK authorities to procure that an arrested person was brought promptly before a judicial officer unless they chose to keep that person in their custody. Thus, Article 5(3) could be complied with by releasing the arrested person if he could not be transferred promptly into the custody of the Afghan authorities. It cannot therefore be said that it was impossible for the UK to secure the right to liberty guaranteed by the Convention. Furthermore, the reasons which constrained the ability of UK forces to keep people lawfully in their custody derived from the inability to fulfil the object of detention operations – which was to deliver those arrested to the Afghan authorities – and not from a lack of extraterritorial power.