Home EJIL Analysis UK Supreme Court Decides R (Smith) v SSD

UK Supreme Court Decides R (Smith) v SSD

Published on June 30, 2010        Author: 

Today the UK Supreme Court decided R (Smith) v Secretary of State for Defence [2010] UKSC 29 (press summary), yet another fascinating addition to the unfolding saga on the extraterritorial application of human rights treaties.

The plaintiff was the mother of a UK soldier stationed in Iraq who died there from a severe heatstroke. She demanded an inquiry into her son’s death that would be compliant with Article 2 ECHR, that would be able to expose what in her view were systemic faults in the UK’s provision of equipment and facilities to its soldiers in Iraq which ultimately led to her son’s death. In other words, the case is a mirror-image of Al-Skeini, which also dealt with Art. 2 procedural obligations in Iraq, but that time with respect to inquiries into the deaths of Iraqi nationals at the hands of UK troops. As the readers are aware, the Grand Chamber of the European Court held hearings in Al-Skeini just a few weeks ago (see my old post for more background).

With regard to extraterritoriality, the issue before the Supreme Court in Smith was this: does a UK soldier in Iraq enjoy the protection of the ECHR while stationed in an area not under the UK’s effective control? Incidentally, on the facts of the case, Private Smith actually died on a UK military base. Per the UK government’s concession in Al-Skeini, the House of Lords’ quite dubious analogy between a military prison or base and an embassy, and the European Court’s recent admissibility decision in Al-Saadoon, that fact alone would have brought Private Smith within the UK’s jurisdiction. Readers will recall that in Al-Saadoon the European Court brought the spatial model of Art. 1 jurisdiction as state effective overall control of a geographical area to its extreme, but saying that a military prison or base qualified as an ‘area’ susceptible to such jurisdiction and control.

In other words, under the spatial model Private Smith would have been within the UK’s jurisdiction, and therefore entitled to protection under Art. 2 ECHR. However, issue was raised in the lower courts as to whether he would have been within the UK’s jurisdiction even if he did NOT die on the base, but in essentially the same circumstances. Like the lower courts, therefore, the Supreme Court was now faced with a set of questions in a quasi-advisory posture – something that several judges openly lamented. The Court nonetheless decided to rule on the matter, because it is one of great practical relevance of UK military operations abroad; Private Smith is obviously not the only UK soldier to have died in Iraq or Afghanistan, and many soldiers lost their lives outside areas under UK effective control.

The lower courts applied to Private Smith a variant of the personal model of Art. 1 jurisdiction, as state authority and control over individuals, finding that he indeed fell within the scope of Art. 1. In their view, simply by virtue of being a part of the UK military, Private Smith was within the UK’s authority and control, and accordingly within its jurisdiction.

Today the Supreme Court disagreed. By a majority of 6 to 3 (Lady Hale and Lords Mance and Kerr dissenting), the justices found that mere membership in the armed forces was insufficient to establish a jurisdictional link for the purposes of Art. 1 ECHR.

For what it’s worth, this result is in my view ultimately correct. I am personally less than happy, however, with the reasoning of either the majority or the minority in the process of reaching this result. Because they as of necessity had to start from Bankovic, a methodologically deeply flawed decision of the European Court, both the judgments of the majority and the minority suffer from a great deal of otherwise needless conceptual confusion. In other words, because they are national judges applying a treaty primarily supervised by an international court, it is natural that the justices will follow the European Court’s approach, however flawed, and whatever their misgivings about the mess that Strasbourg has made. And as in Al-Skeini, several justices thought that this was an issue that was in the end for Strasbourg to settle. Thus, for example, Lord Phillips in para. 60 of his judgment said that:

We are here dealing with the scope of the Convention and exploring principles that apply to all contracting States. The contention that a State’s armed forces, by reason of their personal status, fall within the jurisdiction of the State for the purposes of article 1 is novel. I do not believe that the principles to be derived from the Strasbourg jurisprudence, conflicting as some of them are, clearly demonstrate that the contention is correct. The proper tribunal to resolve this issue is the Strasbourg Court itself, and it will have the opportunity to do so when it considers Al-Skeini. For these reasons I would hold that the Court of Appeal should not have held that Private Smith was within the jurisdiction of the United Kingdom within the meaning of article 1 at times when he was not within premises under the effective control of the army.

Similarly, though in the minority, Lord Mance in paras. 163-164 of his judgment seemed to have had mixed feelings about following Strasbourg. On the one hand it made his life easier, as he did not have to approach Art. 1 ECHR as completely an issue of first impression. On the other, Strasbourg’s wildly conflicting case law prevented him from attaining a totally clear and principled justification for extending ECHR rights to UK soldiers in Iraq. (Incidentally, I am overjoyed that this article of mine was cited by Lord Mance, in the company of such authors as Rick Lawson and former ECtHR Judge Loukis Loucaides, all arguing that the concept of ‘jurisdiction’ in human rights treaties is in reality a simple one, denoting de facto state control over territory and/or individuals, nothing more, and nothing less).

So, Bankovic and its lamentable offspring aside, why is the majority correct that Private Smith was not within the UK’s jurisdiction under the personal model? Isn’t it true, as Lord Mance points out, that it is ‘commonsense’ that UK soldiers are within the UK’s jurisdiction? Isn’t it true, as Lord Kerr thought (para. 322), that ‘the United Kingdom brought its soldiers into Iraq; it not only asserted complete authority over them while they remained there, it explicitly excluded the exercise of authority over those soldiers by any other agency or state; and it has always been clear that soldiers remain subject to the laws of the UK during their service abroad.’ Likewise, isn’t it true that, as Alex Bailin QC pointed out in his case preview at the excellent UK Supreme Court Blog, ‘[t]here is a simple, intuitive point  – if a soldier is always subject to UK law, wherever he may be, should he not thereby be also entitled to the protection of UK law (including the Human Rights Act)?’

I would submit that this intuition is wrong, however sensible it may seem at first glance. It not only assumes that the spatial model can be supplemented by a variant of the personal model based on legal links such as nationality or membership in the armed forces, but it rests on a confusion – confusion that authors such as Lawson, Loucaides, or Ralph Wilde have tried to dispel – between the various meanings that the word ‘jurisdiction’ can have in general international law.

It is true that a state may legislate for its nationals and members of its armed forces even when they are abroad, and in fact all states do that to a greater or a lesser extent. It is also true that a state’s soldiers are its de jure organs, and that their conduct performed in their official capacity is attributable to the state. Neither of these premises can, however, lead to the conclusion that a state’s soldiers, or diplomats, any other agents or even nationals are regardless of their location always within that state’s jurisdiction within the meaning of the jurisdiction clauses of human rights treaties. This ‘jurisdiction’ is simply not the jurisdiction to prescribe in general international law. It is not about the application and extent of UK domestic law.

If the personal model of jurisdiction is valid, its acceptance of nationality or membership in the armed forces as grounds of state jurisdiction would lead to an open embrace of double standards – and there, I may say, is where the intuitive appeal actually lies. Let us, for example, simply transplant Smith to the facts of Al-Skeini – imagine that one of the five applicant Iraqi citizens killed by UK troops on patrol was actually a dual UK/Iraqi national. Further, imagine that the sixth applicant was not Baha Mousa, but was in fact the mother of a UK soldier killed by friendly fire by UK troops on patrol. The reasoning of the lower courts or the minority of the Supreme Court in Smith would lead to the result that the dual national and the UK soldier would be within the UK’s jurisdiction, and thus entitled to the protection of Article 2 ECHR, while the other applicants would not, even though all of them were killed in exactly the same way, in exactly the same place.

In other words, the appeal of the minority approach lies precisely in the fact that we would feel it unjust if the UK government did not owe human rights obligations to its own soldiers in Iraq, while our feelings may be more mixed when it comes to Iraqi civilians shot on patrol. We simply care more about ‘our boys’, then about their enemies or even their victims. The Smith approach is about ‘us’ versus ‘them,’ if not openly so, and is antithetical to the whole idea of universality. This is not to say that nationality or membership in the armed forces are totally irrelevant for the substantive application of human rights norms. Obviously, only a UK national would have the right to vote at any UK election organized abroad, and only a UK national would have the right to return to the UK as his own country. This, however, only depends on the content of a very limited set of rights, but the application of the ECHR as such cannot depend on nationality or any other personal status.

In short, while it may be artificial to say that a UK soldier would be protected by the ECHR while on a UK base, but not off it, this is no more artificial than saying that an Iraqi national is protected by the ECHR while on a UK base or while in UK custody, but not on the streets of Basra, as was the result in Al-Skeini. Thus, though Lord Phillips’ judgment for the majority mainly rests on considerations of practicality or effectiveness to deny the extraterritorial applicability of the ECHR to UK soldiers abroad on account of their status, I would submit that, more fundamentally (and perhaps with a degree of irony), this is the only result consistent with universality as the normative foundation of the ECHR and all other human rights treaties – unless, of course, it is Bankovic and Al-Skeini themselves which are wrongly decided.

(For more commentary, see the UK Human Rights Blog and UK Supreme Court Blog, here and here).

Print Friendly, PDF & Email

14 Responses

  1. Aurel Sari Aurel Sari

    Dear Marko,

    Great first reaction, as always. I would question your position on the exercise of jurisdiction over members of the armed forces, however.

    Sure, States regulate the conduct of their armed personnel when they are abroad just as they regulate the conduct of other officials, including diplomatic agents etc. Nevertheless, the relationship between members of the armed forces and their sending State (which is not necessarily their State of nationality!) is special: it is a relationship of military subordination characterised by the exercise of command and control by military supriors over their military subordinates. The entire system (military orders, superior/subordinate, military discipline etc) is geared towards the exercise of effective control over individuals.

    Being a member of the armed forces is not merely a question of being subject to the regulatory authority of the sending State abroad, but more improtantly for all concerned it is a question being subject to the effective control of that State. For this reason I find it difficult to accept that military personnel on active duty and within an effective chain of command should not be presumed, ipso facto, to fall within the jurisdiction of their sending State for the purposes of Article 1 of the Convention, whether within its territory or abroad.


  2. Marko Milanovic Marko Milanovic

    Hi Aurel,

    Thanks a lot for the comments. I do in fact agree with you to an extent. If we are applying a purely factual test of authority and control over individuals, then I am prepared to accept that a UK soldier could be under such control. But if we are doing that, then we must apply the same test to the Iraqi national who is killed by a UK soldier. What we can’t do is treat them differently (and to my mind at least having the power to kill someone is no less ‘authority and control’ then having a person voluntarily submit to a military chain of command).

    This approach, however, is precluded by Bankovic and Al-Skeini, and that’s what the last sentence of my post tried to say. This is why the justices who favoured the extraterritorial application to Smith did NOT rely on purely factual criteria of control, but on what I would consider to be irrelevancies such as the extension of UK domestic law, exclusion from Iraqi law, mutuality of obligations, etc.

  3. Aurel Sari Aurel Sari

    I take your point about discrimination, but I’m not entirely convinced by it. An alternative reading might focus on the level or effectiveness of control.

    Do you think it is unreasonable to suggest that the exercise of military command and control over individuals represents a more effective form of control than the use of armed force, aerial bombing (Bankovic) or shooting (Al-Skeini), against other individuals? If so, could it be said that the former passes a de minimis test while the latter doesn’t?


  4. Marko Milanovic Marko Milanovic

    We could certainly set some sort of de minimis threshold. The question is whether we could do that based on a non-arbitrary criterion, and I submit that we cannot. For example, if we take a look at Al-Skeini before the Court of Appeal, which actually employed a personal test for jurisdiction, we’ll see that they did indeed set a de minimis threshold, by saying that physical custody qualified as ‘authority and control’ over an individual, thereby covering Baha Mousa who was killed in UK custody, while the power to kill an individual did not qualify as authority and control, thereby rendering the other five applicants outside the UK’s jurisdiction.

    And if we look at the reasoning the Court gave for limiting the personal model to physical custody, we’ll see that the only reason the Court gave was that this resulted was mandated by Bankovic (Brooke LJ, paras. 80-81). That is not a good enough reason, at least in my book.

    Why, after all, should the state’s power to kill an individual not be considered as an exercise of the state’s control over that individual? Is not killing, in fact, the ultimate form of control? It is certainly true that if a state has an individual in custody, it can do more to him than just kill him, by for example subjecting him to torture or inhuman treatment. But should this matter? Aside from killing, one can easily imagine a situation where, for instance, one state has custody, but another does the torture, or is at least complicit in it – as with the alleged involvement of the US and UK intelligence services in interrogations conducted in Syria, Egypt, Pakistan, or what have you. If, say, a CIA agent was to prod a detainee held in Pakistani custody with a hot poker, why should Pakistan be the only state to blame for the torture?

    Indeed, if we take a look at some of the case law, we can see that the personal model has been applied even in cases which did not involve physical custody, as, for example, in Pad, Isaak, and Solomou. All of them involved a killing that took place outside detention. And the whole universality rationale of Issa was that ‘the Convention cannot be interpreted so as to allow a State party to perpetrate violations of the Convention on the territory of another State, which it could not perpetrate on its own territory’ (relying on the Human Rights Committee in Lopez-Burgos).

    From the standpoint of universality, it makes no sense in saying that a state acting extraterritorially may not detain arbitrarily, but that it may, in fact, kill arbitrarily. If anything, such a position would precisely create an incentive for states to kill rather than capture, as can indeed be seen from the escalation of drone attacks against suspected terrorists in Pakistan and Afghanistan, and a massive decrease in the number of those who are actually detained.

  5. @Marko

    What happens if the British army brings its soldiers to an island not yet part of any state (terra nullius), it does not build any camp, leaves the island and then it drops an atomic bomb on them, for fun. Are they or are they not protected by the Article 1 of the Convention?

  6. Marko Milanovic Marko Milanovic


    If the killing takes place in a territory under the effective control of the UK, then Art. 1 ECHR would apply. The House of Lords held that this would only be the case if the territory in question actually belonged to one of the other ECHR states parties, but this is wrong, and the European Court will (probably) say so when it decides Al-Skeini.

    When it comes to a killing in a territory which is not under UK effective control, the practice of the European Court is contradictory. Bankovic says that Art. 1 ECHR wouldn’t apply, but other cases say the opposite. If you ask me what my personal view is what the (case) law should be, then I would say that the negative obligation to respect human rights applies to all state conduct – i.e. every killing would be covered by the ECHR, no matter where it takes place – but that the positive obligation to secure human rights would apply only to areas under state effective overall control.

  7. @Marko
    And thus we come to my mantra that I repeat like a broken record: there are two major flaws in the international law, one is that compulsory jurisdiction is almost unexisting and the second that individuals cannot sue states at international courts. We should create one or more international courts in each state, paid by the state but considered independent and international, for instance U.N. courts. In those courts the individual should be able to sue any state, her own or foreign. And the applied law should be in the first instance the non-derogable articles in ICCPR. The decisions could be appealed at regional courts (still to be created) and the decisions of the regional courts could be appealed at ICJ. Article 34 could be amended to make this possible. This would solve many of the contradictions in IL. For instance if the spouse of one of the target-killed guy in Somalia doubts the U.S. legal argument, she could sue U.S. there and the judges would decide which of the parties is right.

  8. Aurel Sari Aurel Sari

    I fully agree with you Marko that setting a minimum threshold of control is difficult, at least it is difficult to do so in a non-arbitrary way. Handcuffing someone qualifies, shooting does not – as you say, depriving an individual of their life could be considered an ultimate form of control and it is difficult to argue, morally and perhaps also logically, why it should constitute control for the purposes of Article 1.

    Nonetheless, I think there is a danger of taking this too far. First, in a situation such as Bankovic, I think it stresses the notion of control to suggest the States exercise control over individuals who are killed or hurt by way of aerial bombardment. Unless the individual was targeted directly (i.e. targeted killing style), I just do not see how the States could be said to exercise control in a factual sense – the bomb could hit me, you or our neighbour down the street. Are we all within the control of that State? Certainly, we could say that we do, but in that case we would no longer be talking about control in a truly factual, effective sense, but more as a legal construct that allows us to establish a connection between the individual and the State concerned. This is partly why, secondly, I think that a notion of control that is not based on fact and some level of effectiveness comes dangerously close to replicating the notion of State jurisdiction in the sense of the exercise of State regulatory power. If falling under the control of the State for the purposes of Article 1 simply means being affected by the action of that State, even incidentally, why should the mere existence of jurisdictional competence by that State (eg over nationals abroad) not suffice to trigger the application of Article 1?

    In short, if we are to distinguish jurisdiction as effective control from jurisdiction as State regulatory authority, as you and others have argued, we must insist in some level of effectiveness and factual control otherwise the distinction is lost.


  9. Marko Milanovic Marko Milanovic


    I think you are quite right to point out the dangers of an expansive approach to extraterritorial application. As a matter of policy, do we really want the European Court to be the final arbiter on any single killing by European troops in Afghanistan or Iraq or what have you? Is it really institutionally competent to do so? Does it have access to a reliable factual record? Can it interpret the ECHR with a sufficient degree of flexibility, for example by taking into account the rules of IHL, in order to make the applicability of the ECHR to such extraordinary circumstances realistic rather than utopian? And would doing so actually have the potential to compromise the integrity of the ECHR regime, both in terms of practicality (e.g. opening the floodgates to litigation for an already overwhelmed court), and in terms of values (e.g. by allowing preventive detention, or having a more relaxed necessity standard for killing).

    My own view is that all of confusions in the case law etc. are ultimately explicable by reference to this policy tension between universality and effectiveness. Where that tension is best addressed, however, is on the merits of substantively very difficult and complex cases. And because courts are reluctant to address these merits, they fashion wholly artificial limitations on the territorial scope of human rights treaties (or for that matter, domestic constitutional protections of individuals rights), only to carve out even more artificial exceptions from these limitations when it would be morally intolerable not to do so (e.g. Baha Mousa in Al-Skeini is protected by the ECHR because a military prison is somehow analogous to an embassy).

    Coming back to your last comment, I would say that it presupposes that we actually need to have a threshold for violations of negative obligations. I would say that no such threshold exists, or if it is imposed, it will invariably tend to collapse, because the normative pull of universality is such that it is hard to rationalize why a state which is fully in control of the conduct of its own organs and agents should not have the obligation to refrain from violating the human rights of others, when it is perfectly capable of complying with this obligation. In other words, I would argue that the ‘jurisdiction’ threshold in Art. 1 ECHR applies (as the text says!) only to the obligation to SECURE human rights, but not to the implicit obligation to RESPECT them.

    If, on the other hand, we do use a threshold – that of a personal model of jurisdiction as authority and control over an individual – then I would say that practically ANY exercise of such authority and control would suffice to satisfy it. For example, if a state puts a fugitive who has absconded to another country on trial in absentia, that would to my mind satisfy the factual jurisdiction threshold, even though the state has not touched the individual physically (and we do indeed have such cases, in which nobody has raised an Art. 1 ECHR issue!).

    In other words, my factual conception of jurisdiction as state power over an individual would encompass the exercise of a LEGAL power (such as e.g. the issuance of an arrest warrant – see, e.g., Stephens v. Malta, or the taking of domestic property of a foreign individual). However, the exercise of that power need not be LAWFUL in international law – e.g. it might or might not constitute the violation of the sovereignty of the territorial state. In that sense the Art. 1 notion of jurisdiction would be distinct from the concept of jurisdiction to prescribe in general international law.

  10. @Marko

    ==As a matter of policy, do we really want the European Court to be the final arbiter on any single killing by European troops in Afghanistan or Iraq or what have you?==

    If someone kills me I definitely want that there is a final arbiter of the legality of the fact. There is no good reason why the one that kills me should have the last word, nor that she is the final arbiter.

    ==Does it have access to a reliable factual record?==

    If someone kills me the only thing I have to prove is that I was killed. Then the burden of proof shifts to the killer to prove that she had a good reason for doing it. If she cannot prove that than she is liable. The same should be here the case. If the state doing the killing cannot prove that it had a good reason, than it should be liable. Precicely like in Loizidou v. Turkey.

  11. Aurel Sari Aurel Sari

    Thanks for your detailed replies Marko. I think you make an important distinction here between the exercise of control over an individual by the State through legal processes or institutions and the scope of the State’s regulatory competence under international law. As I understand it, that distinction is important because the mere fact that an individual falls within the regulatory competence of the State does not necessarily mean that Article 1 ECHR is applicable (State jurisdiction qua competence does not necessarily equal jurisdiction qua control for the purposes of Article 1), while the State may use legal processes and institutions to exercise control over an individual even though those processes and institutions are not based on or subject to its regulatory competence (eg acting abroad without the consent of the territorial State or instituting measures under a different legal system). I’m quite happy to leave it at that.


  12. […] Read more comments on the judgment:Adam Wagner, UK Human Rights BlogMarko Milanovic, EJIL Talk. […]

  13. Hanne Garberg

    Dear Sir,

    Very interesting comment. Without having read all the replies, I cannot completely understand your argument of why the application of ECHR cannot depend on nationality? The fact that the UK is a member of the Council of Europe and is sending one of their own nationals abroad under UK command, should this not be enough for the soldier to fall with in the UK’s jurisdiction?

    Even though it might be unfair that an Iraqi killed at the same place at the same time will not be covered by the Convention, I fail to see why this is relevant since the ECHR mainly is an obligation for the State towards its own nationals.

    I do also think it is more artificial to say that an Iraqi national is protected while in UK custody but not on the street of Basra, than it is to say that a UK soldier would be protected by the ECHR on a base but not off it. It is clear that the UK cannot be responsible for every Iraqi or other nationals in Iraq, but I cannot see that this should affect the UK’s responsibility regarding a soldier who is acting on behalf of his state.

    Kindest regards,
    Hanne Garberg.

  14. […] off the cuff, is not so easy. But essentially, the Guantanamo point is in parallel in some ways to the Jason Smith case, which was an inquest case about whether the UK has jurisdiction over bases in Iraq, and so on. […]