On 21 January the Palestinian Minister of Justice lodged with the ICC Registrar a ‘Declaration Recognizing the Jurisdiction of the International Criminal Court’ over acts committed on the territory of Palestine since 2002. This article concerns three issues regarding the admissibility of this declaration, all of which are linked to the question of statehood. It first argues that the ICC Prosecutor may not assume the existence of a Palestinian state because the Palestinians themselves do not make a claim to that effect. It then examines whether under a purposive interpretation of Article 12(3), declarations should also be admitted from quasi-states. It argues that there is no justification for such a purposive interpretation, as the ICC Statute already contains an adequate mechanism to deal with exceptional situations such as that of the Gaza Strip. Finally, the article examines the consequences of the ICC Prosecutor engaging in questions concerning statehood and recognition. It argues that taking on the Palestine case would open up a Pandora’s Box and risk turning the ICC into a political playing field for aspirant entities in search of international status.

For what it’s worth, I am personally in broad agreement with Yael’s argument, even if I don’t think that it is in principle objectionable for the ICC or for its prosecutor to address issues of statehood purely as preliminary questions. For a different take than Yael’s, Alain Pellet has prepared a legal opinion arguing that the term ’state’ in Article 12(3) should be interpreted purposefully so as to allow the Palestinian declaration, even if Palestine is not a state as a matter of general international law – see more here on Bill Schabas’ blog. We’ll see, of course, what the Court ultimately makes of the whole thing.

Many congratulations to Andreas! He is sure to make a strong contribution at the Court, where his international law expertise may prove to be invaluable. And we might just get him to blog for us again sometime in the future.

 Ben Batros is Appeals Counsel at the International Criminal Court and a former Senior Legal Officer, International Crime Branch, Australian Attorney-General’s Department. Philippa Webb, is Visiting Assistant Professor at Leiden University and former Special Assistant to President of the International Court of Justice

Last week (on 25 February), the Australian Federal Court handed down its decision in Habib v. Commonwealth of Australia ([2010] FCAFC 12).  In some respects, this case is part of several emerging trends relating to immunities and human rights. First, it concerns allegations of torture by foreign State officials, including in Guantanamo Bay (see previous EJIL:Talk! posts here and here)   and  Second, it involves an alleged victim claiming damages against a Government, which resonates with the practice in US courts under the Alien Torts Claims Act and the Torture Victim Protection Act (including the Samantar case being heard by the US Supreme Court this week – see post by Julian Ku over on Opinio Juris). Third, a Government has asked the court to throw out the case, arguing that the courts of one country cannot sit in judgment on the acts of the government of another – echoes of Al-Adsani v. United Kingdom at the ECtHR, Jones v. Saudi Arabia in the UK House of Lords, and the pending Germany v. Italy case at the ICJ.

But even though Mr Habib’s claim lies at the intersection of these trends, it does not follow the “typical” model of a civil claim against the agents directly responsible for the alleged abuses; nor does it pursue a criminal prosecution, where questions of immunity have recently formed a less imposing barrier.  Rather, Mr Habib’s lawyers appear to have considered the experience of similar claims abroad, and to have crafted a case which takes into account both the particular opportunities and obstacles presented by Australia’s legislative regime. As a result, the heart of dispute is the scope of the act of state doctrine, not the applicability of state immunity.

Mr. Habib, an Australian citizen, was detained by Pakistani authorities in early October 2001, and transferred to Egypt in November 2001, to Bagram airbase in Afghanistan in April or May 2002, and then to Guantanamo Bay where he was detained until his release (without charge) in January 2005.  During these periods, Mr Habib alleged a range of mistreatments amounting to torture and inhumane treatment which Australian federal criminal law criminalises with extraterritorial (and in most cases universal) jurisdiction (paras. 3, 15-21, 57-65 of the decision).  The Criminal Code also makes it an offence to aid, abet or counsel the commission of a crime.  Mr Habib’s allegations included that Australian police, intelligence and diplomatic officials had visited him during his detention (which was not disputed), been present during questioning, and may even have provided information used in the interrogation (paras. 17-18, 66-67), and the facts if proved might thus found criminal cases against both the foreign agents who conducted the interrogations and the Australian officials who assisted them. 

To circumvent certain procedural requirements for prosecution of these particular crimes, Mr Habib’s lawyers brought a civil action.  However, with the exception of claims under the Alien Torts Claims Act in the US, civil claims for alleged torture at the hands of foreign state officials have previously been held to be barred by state immunity (notably in the UK in Jones v Saudi Arabia), as Mr Habib’s counsel expressly conceded.  So Mr Habib sued the Australian Government for the torts of misfeasance in public office and of intentional infliction of indirect harm by its officials when they aided and abetted his alleged torture.

An action based on aiding and abetting does not require that the principal has been convicted, or even prosecuted.  Yet to prove that any Australian official had acted beyond their authority and therefore in misfeasance of office by aiding and abetting torture required proof (in this case to the civil standard) that the crimes had been committed by the foreign officials.  There was thus a duality at the core of Mr Habib’s claim – on one hand, an Australian citizen sought redress from the Australian Government under Australian (procedural) law for alleged acts of Australian officials prohibited under Australian (substantive) law; but to resolve this, an Australian court would have to determine whether acts of foreign officials outside of Australia constituted crimes under Australian (extraterritorial) law (paras. 68-71, also 22-23). 

The decision addressed a discrete legal issue which asked the Federal Court to balance these two aspects – is the claim non-justiciable under the act of state doctrine because its resolution would require a determination of the unlawfulness of acts of agents of foreign states within their territory?  The Court answered “no” for varying reasons.

For Perram J, this question of balance was decisive:  the Court’s constitutional obligation to review whether Australian officials had acted beyond their authority under Commonwealth law trumped any common law principle of act of state (para. 24).  Jagot J, on the other hand, prioritised analysis of act of state, spending over half of her opinion discussing its origins, development, and the considerations informing the content of the doctrine (paras. 72-120).  Jagot J (with whom Black CJ agreed) was clear that the act of state doctrine cannot preclude the Court from making findings of fact regarding Mr Habib’s treatment at the hands of foreign officials when those officials are not subject to the jurisdiction of the Court.  Indeed, the Chief Justice emphasised that this is not a question where the authorities were finely balanced; but that even that if it was, the Court should give effect to the Parliament’s moral choice proscribing torture as offensive to the ideals of humanity and a crime even where committed outside of Australia and in an official capacity (paras. 7-11). 

Jagot J’s finding that the act of state doctrine did not bar consideration of this case was supported by the settled and consensual nature of international law prohibiting torture, and the clear judicial standards provided by the Australian legislation in question (paras. 108, 117-120).  And ultimately, in light of her assumption that state immunity would prevent an action against a foreign official, Jagot J appeared to be swayed by the logic that applying the act of state doctrine in this case would mean that the Australian officials could not be held accountable in any court (paras. 113-114).  Even without Perram J’s fixation on the Court’s constitutional role, this would have been a bridge too far.

In the Justices’ discussion of and engagement with decisions by a wide range of national and international courts, this case demonstrated the value and importance of judicial dialogue. However, while the decision is clear about what act of state does not cover, it does little to clarify the precise nature and content of the doctrine:  Perram J considered that “[b]eyond the certainty that the doctrine exists there is little clarity as to what constitutes it” (para. 38); and Jagot J referred repeatedly to authorities and submissions attesting to its “uncertain” nature and application (paras. 51-52).  This vagueness perhaps means that Habib v. Cth will not easily be drawn upon by courts in other jurisdictions engaged in analysing the act of state doctrine.

The future applicability of this case in other jurisdictions is also limited by the fact that the claim was brought in an Australian court against Australian officials:  it constitutes a rare example of human rights litigation involving act of state but not the applicability of state immunity. Nonetheless, the Federal Court still commented on state immunity, appearing to assume that foreign officials could successfully claim state immunity if sued in an Australian court and Australian officials if sued in a foreign court (para. 113).  Although this was obiter dicta, as this question was not before it, the Federal Court seems to have positioned itself alongside the House of Lords in Jones and against any exception to state immunity for jus cogens violations (at least in relation to civil claims), even before the ICJ judgment pronounces on this very question in Germany v Italy (see earlier post here).

Readers will recall that the basic question raised by the case is whether the transfer by the UK of the applicants who were in the custody of UK troops in Iraq to Iraqi authorities for trial violated the applicants ECHR rights, specifically the non-refoulement principle established by the Court in Soering v. UK, inter alia because there was serious risk of them being subjected to the death penalty. In Soering itself the issue was the surrender of the applicant to the US, where there was serious risk of him being subjected to the death penalty. However, the death penalty was at the time still not outlawed with respect to the UK by Protocols 6 and 13, and so the actual issue was inhuman treatment that the applicant would suffer as a consequence of the death row phenomenon. Likewise, without the two protocols, Article 2(1) ECHR specifically contemplates the death penalty, and it as such could not be held to be contrary to other provisions of the Convention, namely Article 3 prohibiting all forms of ill-treatment.

Now in Al-Saadoon we have the first merits judgment dealing specifically only with the death penalty and non-refoulement. The Court started its analysis by considering the developments in both treaty action and in its case law with regard to the death penalty since Soering:

118. The Court considers that, in respect of those States which are bound by it, the right under Article 1 of Protocol No. 13 not to be subjected to the death penalty, which admits of no derogation and applies in all circumstances, ranks along with the rights in Articles 2 and 3 as a fundamental right, enshrining one of the basic values of the democratic societies making up the Council of Europe. As such, its provisions must be strictly construed (see, mutatis mutandis, Soering, cited above, § 88; McCann and Others v. the United Kingdom, judgment of 27 September 1995, § 147, Series A no. 324).

…

120. It can be seen, therefore, that the Grand Chamber in Öcalan did not exclude that Article 2 had already been amended so as to remove the exception permitting the death penalty. Moreover, as noted above, the position has evolved since then. All but two of the Member States have now signed Protocol No. 13 and all but three of the States which have signed have ratified it. These figures, together with consistent State practice in observing the moratorium on capital punishment, are strongly indicative that Article 2 has been amended so as to prohibit the death penalty in all circumstances. Against this background, the Court does not consider that the wording of the second sentence of Article 2 § 1 continues to act as a bar to its interpreting the words “inhuman or degrading treatment or punishment” in Article 3 as including the death penalty (cf. Soering, cited above, §§ 102-104).

123. The Court further reiterates that expulsion by a Contracting State may give rise to an issue under Article 3, and hence engage the responsibility of that State under the Convention, where substantial grounds have been shown for believing that the person concerned, if deported, faces a real risk of being subjected to treatment contrary to Article 3. In such a case Article 3 implies an obligation not to deport the person in question to that country (see Saadi v. Italy, cited above, § 125). Similarly, Article 2 of the Convention and Article 1 of Protocol No. 13 prohibit the extradition or deportation of an individual to another State where substantial grounds have been shown for believing that he or she would face a real risk of being subjected to the death penalty there (see Hakizimana v. Sweden (dec.), no. 37913/05, 27 March 2008; and see, mutatis mutandis, Soering, cited above, § 111; S.R. v. Sweden (dec.), no. 62806/00, 23 April 2002; Ismaili v. Germany (dec.), no. 58128/00, 15 March 2001; Bader and Kanbor, cited above, § 42; Kaboulov v. Ukraine, no. 41015/04, § 99, 19 November 2009).

The Court has now clearly held that Article 2 ECHR provides for a non-refoulement obligation: there can be no extradition if a serious risk of the death penalty is established, period. Recall, however, the UK government’s argument that the applicants in this case were not held in the UK, but in Iraq, and that the UK had a legal obligation to transfer them to Iraqi authorities. In the UK’s view, this necessitates that the non-refoulement obligation be qualified to take into account these exceptional circumstances, and the fact that the UK would otherwise be forced to violate its obligations towards Iraq. (Such was also the approach of the Court of Appeal in R (B) v Secretary of State for Foreign and Commonwealth Affairs ([2004] EWCA Civ 1344).

This is how the Court now approached the matter:

126. The Government contended that they were under an obligation under international law to surrender the applicants to the Iraqi authorities. In this connection, the Court recalls that the Convention must be interpreted in the light of the rules set out in the Vienna Convention on the Law of Treaties, 1969, of which Article 31 § 3(c) indicates that account is to be taken of “any relevant rules of international law applicable in the relations between the parties”. More generally, the Court recalls that the principles underlying the Convention cannot be interpreted and applied in a vacuum. The Convention should be interpreted as far as possible in harmony with other principles of international law of which it forms part (see Al-Adsani v. the United Kingdom [GC], no. 35763/97, § 55, ECHR 2001-XI; Banković, cited above, §§ 55-57). The Court has also long recognised the importance of international cooperation (see Al-Adsani, § 54 and Bosphorus, § 150, both cited above).

127. The Court must in addition have regard to the special character of the Convention as a treaty for the collective enforcement of human rights and fundamental freedoms. Its approach must be guided by the fact that the object and purpose of the Convention as an instrument for the protection of individual human beings requires that its provisions be interpreted and applied so as to make its safeguards practical and effective (see, inter alia, Soering, cited above, § 87; Loizidou v. Turkey (preliminary objections), cited above, § 72; McCann and Others, cited above, § 146).

128. It has been accepted that a Contracting Party is responsible under Article 1 of the Convention for all acts and omissions of its organs regardless of whether the act or omission in question was a consequence of domestic law or of the necessity to comply with international legal obligations. Article 1 makes no distinction as to the type of rule or measure concerned and does not exclude any part of a Contracting Party’s “jurisdiction” from scrutiny under the Convention (Bosphorus, cited above, § 153). The State is considered to retain Convention liability in respect of treaty commitments subsequent to the entry into force of the Convention (see Bosphorus, cited above, § 154 and the cases cited therein). For example, in Soering, cited above, the obligation under Article 3 of the Convention not to surrender a fugitive to another State where there were substantial grounds for believing that he would be in danger of being subjected to torture or inhuman or degrading treatment or punishment was held to override the United Kingdom’s obligations under the Extradition Treaty it had concluded with the United States in 1972.

These three paragraphs are crucial in explaining the Court’s approach to possible norm conflicts. Witness, first, how it blows both hot and cold. In para. 126 it allows for the possibility of taking into account other international obligations when interpreting the ECHR, in order to avoid possible conflict. This, of course, what the Court actually did in both Al-Adsani and in Bosphorus. Then in para. 127 it however reiterates the importance of the ECHR regime, implying that it does not avoid norm conflicts lightly. Finally, in para. 128 the Court quite rightly points out that a norm conflict also existed in Soering between the UK’s obligations under the ECHR and its obligations under the US extradition treaty, but that this was not grounds enough for reading down the ECHR.

My only quibble is with the Court’s statement that the ECHR obligations ‘was held to override’ the UK’s obligations towards the US. This is simply untrue, at least if the word ‘override’ is to be taken as implying a hierarchical relationship. The ECHR is certainly not superior law to a bilateral treaty with a non-state party. Rather, the two treaties were in an unresolvable norm conflict, with the only possible solution to that conflict being political in nature, for example a renegotiation of either one – see more here, esp. at 16-21.

Having now outlined its approach to norm conflicts, and having held that Articles 2 and 3 ECHR now generated a non-refoulement obligation with respect to the death penalty as such, the issue to be decided was whether this obligation should apply any differently to UK troops in Iraq. The Court said no:

137. Protocol No. 13 came into force in respect of the United Kingdom on 1 February 2004. The Court considers that, from that date at the latest, the respondent State’s obligations under Article 2 of the Convention and Article 1 of Protocol No. 13 dictated that it should not enter into any arrangement or agreement which involved it in detaining individuals with a view to transferring them to stand trial on capital charges or in any other way subjecting individuals within its jurisdiction to a real risk of being sentenced to the death penalty and executed. Moreover, it considers that the applicants’ well-founded fear of being executed by the Iraqi authorities during the period May 2006 to July 2009 must have given rise to a significant degree of mental suffering and that to subject them to such suffering constituted inhuman treatment within the meaning of Article 3 of the Convention.

138. The Government have contended that, in accordance with well established principles of international law, they had no option but to respect Iraqi sovereignty and transfer the applicants, who were Iraqi nationals held on Iraqi territory, to the custody of the Iraqi courts when so requested. In this respect, however, the Court refers to its case-law, summarised in paragraphs 126-128 above, to the effect that it is not open to a Contracting State to enter into an agreement with another State which conflicts with its obligations under the Convention. This principle carries all the more force in the present case given the absolute and fundamental nature of the right not to be subjected to the death penalty and the grave and irreversible harm risked by the applicants.

139. The domestic courts considered themselves bound by the principles of international law concerning “diplomatic asylum”, as applied by the Court of Appeal in the R(B) case (see paragraphs 58, 72 and 94 above), to find that the duty to provide refuge extra-territorially could operate only where there was clear evidence that the territorial State intended to subject the individual to treatment so harsh as to constitute a crime against humanity. It is not necessary in this judgment for the Court to examine generally the principles of “diplomatic asylum” or to establish when, if ever, the surrender of an individual by a Contracting State’s diplomatic or consular agents could give rise to a violation of the Convention. It merely notes in passing that the Commission in its admissibility decision in W.M. v. Denmark, no. 17392/90, Commission decision of 14 October 1992, Decisions and Reports 73, p. 193, appeared to assume, albeit without detailed reasoning, that the Soering principle against refoulement would apply where an individual sought and was refused refuge in a Contracting State’s embassy.

140. The Court considers in any event that the facts of the present case are such as clearly to distinguish it from a situation of “diplomatic asylum”, for the following reasons. Diplomatic and consular premises have a particular status under international law. When a State sets up a diplomatic mission it agrees to respect the laws of the territorial State and not to interfere in its internal affairs (Vienna Convention on Diplomatic Relations, Article 41 § 1: see paragraph 93 above); this is one of the conditions on which the territorial State consents to the establishment of the mission. Thus, when an individual seeks refuge at an embassy, the obligations owed by the sending State to the territorial State are known and apply ab initio (although there may be other conflicting obligations, for example under the Convention). In contrast, in the present case, the applicants did not choose to seek refuge with the authorities of the United Kingdom; instead, the respondent State’s armed forces, having entered Iraq, took active steps to bring the applicants within the United Kingdom’s jurisdiction, by arresting them and holding them in British-run detention facilities (see Al-Saadoon and Mufdhi v. the United Kingdom (dec.), no. 61498/08, §§ 84-89, 30 June 2009). In these circumstances, the Court considers that the respondent State was under a paramount obligation to ensure that the arrest and detention did not end in a manner which would breach the applicants’ rights under Articles 2 and 3 of the Convention and Article 1 of Protocol No. 13.

141. In any event, the Government have not satisfied the Court that the need to secure the applicants’ rights under Articles 2 and 3 of the Convention and Article 1 of Protocol No. 13 inevitably required them to act in contravention of Iraqi sovereignty. It does not appear from the evidence before the Court that, despite the concerns voiced at ministerial level as early as July 2004 about the risk of the imposition of the death penalty if the applicants were tried by the Iraqi courts (see paragraph 44 above), any real attempt was made to negotiate with the Iraqi authorities to prevent it. According to the evidence of Mr Watkins before the Divisional Court and Court of Appeal (see paragraphs 56 and 66 above), it was the judgment of the United Kingdom Government towards the end of 2008 that it would not be politic even to raise with the Iraqi Government the possibility of removing the applicants to the United Kingdom or continuing to detain them in Iraq after 31 December 2008. However, it would appear from the minute of the DIRC meeting of 28 September 2004 that the Iraqi prosecutors initially had “cold feet” about bringing the case themselves, because the matter was “so high profile” (see paragraph 44 above). This could have provided an opportunity to seek the consent of the Iraqi Government to an alternative arrangement involving, for example, the applicants being tried by a United Kingdom court, either in Iraq or in the United Kingdom. It does not appear that any such solution was ever sought.

142. The Government accept, moreover, that no attempt was made, during the negotiations for the United Kingdom-Iraqi MoU of November 2004 or at any other time, to seek a general assurance from the Iraqi authorities that, in the light of the United Kingdom’s binding obligations under the Convention and Protocol No. 13, no individual transferred from the physical custody of the United Kingdom Armed Forces could be subjected to the death penalty. Similarly, the Government do not contend that, before the decision was made to refer the applicants’ cases to the Iraqi courts, any request was made to the Iraqi authorities for a binding assurance that, if the cases were referred, the applicants would not be at risk of capital punishment. Indeed, it would appear that it was only after the applicants had lodged an application for judicial review before the Divisional Court that a first effort was made to seek clemency on their behalf. However, as the domestic courts found, no binding guarantee was obtained.

143. In summary, therefore, the Court considers that, in the absence of any such binding assurance, the referral of the applicants’ cases to the Iraqi courts and their physical transfer to the custody of the Iraqi authorities failed to take proper account of the United Kingdom’s obligations under Articles 2 and 3 of the Convention and Article 1 of Protocol No. 13 since, throughout the period in question, there were substantial grounds for believing that the applicants would face a real risk of being sentenced to death and executed.

144. The outcome of the applicants’ case before the IHT is currently uncertain. While the applicants remain at real risk of execution since their case has been remitted for reinvestigation, it cannot at the present time be predicted whether or not they will be retried on charges carrying the death penalty, convicted, sentenced to death and executed. Whatever the eventual result, however, it is the case that through the actions and inaction of the United Kingdom authorities the applicants have been subjected, since at least May 2006, to the fear of execution by the Iraqi authorities. The Court has held above that causing the applicants psychological suffering of this nature and degree constituted inhuman treatment. It follows that there has been a violation of Article 3 of the Convention.

145. In the circumstances, and in view of the above finding, the Court does not consider it necessary to decide whether there have also been violations of the applicants’ rights under Article 2 of the Convention and Article 1 of Protocol No. 13.

The Court’s judgment is in my view exactly right. But it bears noting how the legal judgment actually depends entirely on an underlying value judgment as to the importance and nature of Article 2 and 3 rights, and the non-refoulement obligation which protects them. It is also interesting to see how the Court attempts to somewhat narrow down its holding, though I think not entirely persuasively. First, para. 137 could be taken as implying some sort of lex prior, that the Convention only ‘prohibits’ the conclusion of contrary agreements which post-date it. This is not necessarily the case. A century-old extradition treaty, for example, could just as easily run afoul of the ECHR, and the result should in my judgment be no different. Second, the Court’s discussion of diplomatic asylum in paras. 139-140 might perhaps suggest that a different approach should be warranted in an embassy case. Again, however, the fact that a state’s conflicting obligations might be clearer does not necessarily make them more important.

What I do find somewhat strange is the Court’s ruling that because there was a violation of Article 3, there was no need for it to decide whether there was a violation of Article 2 and Protocol 13 (paras. 144-145). Bearing in mind all that it said with regard to non-refoulement under Article 2, the finding of a violation under Article 3 necessarily implies a violation of Article 2. Perhaps I am missing something, but I don’t see how it could be the case otherwise.

The Court then moved to non-refoulement under Article 6, but did not find a violation, holding that no serious risk of a flagrant denial of a fair trial in Iraq was established (paras. 149-50). More interestingly, it then examined the UK’s refusal to obey the interim measures that it had indicated:

162. As stated above, the Court’s approach in interpreting the Convention must be guided by the fact that its object and purpose as an instrument for the protection of individual human beings requires that its provisions be interpreted and applied so as to make its safeguards practical and effective. It has found that the decisions of the United Kingdom authorities to refer the applicants’ cases to the Iraqi courts in December 2005 and to transfer them physically to Iraqi custody on 31 December 2008, without having first received any binding assurance that they would not be subjected to the death penalty, put them at real risk of being executed. It has further found that, as a matter of principle, it was not open to the respondent State to enter into an agreement or arrangement with another State which conflicted with its obligations under Articles 2 and 3 of the Convention and Article 1 of Protocol No. 13. Finally, it has found that the Government have failed to establish that there were no realistic or practicable means available to them by which to safeguard the applicants’ fundamental human rights. In these circumstances, the “objective impediment” claimed by the Government, namely the absence, on 31 December 2008, of any available course of action consistent with respect for Iraqi sovereignty other than the transfer of the applicants, was of the respondent State’s own making.

163. Moreover, the Government have not satisfied the Court that they took all reasonable steps, or indeed any steps, to seek to comply with the Rule 39 indication. They have not informed the Court, for example, of any attempt to explain the situation to the Iraqi authorities and to reach a temporary solution which would have safeguarded the applicants’ rights until the Court had completed its examination.

164. It is true that from June 2008 the Government began to make approaches to the Iraqi authorities, initially to ascertain from the President of the IHT what steps would be effective under Iraqi law and practice to reduce the risk of the applicants receiving the death sentence (see paragraphs 50-54 above). Subsequently, in July 2008 the Government contacted the Iraqi prosecutors and requested them to inform the IHT that the Government were opposed to the death penalty and that neither of the families of the two murdered soldiers wished it to be imposed (see paragraph 86 above). These contacts may or may not have contributed to the IHT’s decisions to amend the charges against the applicants (see paragraphs 86-88 above); this is not a matter on which the Court is able to speculate. However, as the domestic courts also found, the Government’s approaches to the Iraqi authorities prior to the transfer of the applicants on 31 December 2008 were not sufficient to secure any binding assurance that the death penalty would not be applied and the applicants remained at real risk when the United Kingdom decided not to comply with the Court’s indication under Rule 39. The Government’s efforts in 2009 to persuade the Iraqi authorities not to use the death penalty came after the applicants had left United Kingdom jurisdiction and therefore at a time when the United Kingdom authorities had lost any real and certain power to secure their safety. Moreover, to date no binding assurance that the death penalty will not be applied has been given and the applicants’ fate remains uncertain.

165. In conclusion, the Court does not consider that the authorities of the Contracting State took all steps which could reasonably have been taken in order to comply with the interim measure taken by the Court. The failure to comply with the interim measure and the transfer of the applicants out of the United Kingdom’s jurisdiction exposed them to a serious risk of grave and irreparable harm.

166. In the admissibility decision of 30 June 2009 (see paragraph 7 above) the Court joined the question of the admissibility of Article 13 to the merits. Having now examined the merits, and having found in connection with Article 34 that there was no objective justification for the transfer, it reaches the conclusion on similar grounds that the effectiveness of any appeal to the House of Lords was unjustifiably nullified as a result of the Government’s transfer of the applicants to the Iraqi authorities. The Court therefore finds the complaint under Article 13 admissible and it finds violations of Articles 13 and 34 of the Convention.

So far the judgment was unanimous, but here Judge Bratza partially dissented:

The majority of the Chamber have found the Government’s non-compliance with the Rule 39 indication to be unjustified on two principal grounds. It is said that there was no “objective impediment” to compliance with the interim measures since the absence on 31 December 2008 of any available course of action consistent with respect for Iraqi sovereignty other than the transfer of the applicants was of the respondent State’s own making. Secondly, it is said that the Government have not satisfied the Court that they took all reasonable steps, or indeed any steps, to seek to comply with the Rule 39 indication, not having informed the Court of any attempt to explain the situation to the Iraqi authorities or to reach a temporary solution which would have safeguarded the applicants’ rights until the Court had completed its examination.

I am not convinced by either of these points. As to the former, the question whether there was an objective impediment to compliance with an interim measure must be assessed at the time when the measure was applied, in this case 30 December 2008. At that time there existed, as the Court of Appeal found, an objective legal impediment to continuing to detain the applicants and refusing to surrender them to the Iraqi authorities. The fact that, had the United Kingdom obtained the necessary assurances from those authorities some four years before, the applicants could have been safely transferred in December 2008, while undoubtedly relevant in the context of the complaint under Article 3 of the Convention, does not in my view affect the question which falls to be examined under Article 34. As to the latter point, while there are strong reasons to believe that the relevant assurances could have been obtained before the referral of the applicants’ case to the Iraqi courts, the lack of success of the efforts made after June 2008 would clearly suggest that there was no realistic prospect of obtaining such assurances or achieving a temporary solution at a time when the expiry of the Mandate was imminent, a point confirmed by the evidence of Mr Watkins before the Divisional Court and the Court of Appeal (see paragraphs 56 and 66).

For these reasons, while agreeing that in the circumstances of this case the surrender of the applicants violated their rights under Article 3, I would not find it either necessary or appropriate to hold that there had additionally been a violation of Article 34 of the Convention.

With the greatest respect for Judge Bratza, the majority does have it right here. Yes, the UK had a legal obligation to comply with the Iraqi request – but it still had an ECHR obligation to reject it. As the Chamber rightly pointed out, this norm conflict was entirely of the UK’s own making. It was free to choose which obligation to honour, and which it violate. It made its choice, and that choice invariably ran afoul of the Court’s interim measures. It is only if Article 34 was interpreted as creating a justification for non-compliance in cases of potential norm conflict that the result could be otherwise – but I see no reason why Article 34 should be interpreted that way

Finally, as for remedies, the Court ordered the UK government as follows:

171. In the present case, the Court has found that through the actions and inaction of the United Kingdom authorities the applicants have been subjected to mental suffering caused by the fear of execution amounting to inhuman treatment within the meaning of Article 3. While the outcome of the proceedings before the IHT remains uncertain, that suffering continues. For the Court, compliance with their obligations under Article 3 of the Convention requires the Government to seek to put an end to the applicants’ suffering as soon as possible, by taking all possible steps to obtain an assurance from the Iraqi authorities that they will not be subjected to the death penalty.

However, though it did award the applicants costs, it did not award them any damages, ruling that a finding of violation was sufficient just satisfaction. Bearing in mind the magnitude of the violations, this is perhaps not entirely satisfactory – yet this is what the Court frequently does in order to soften the blow somewhat for the government, particularly when the applicants themselves are probably some quite nasty, nasty people.

In sum, this was a judgment well worth the wait. We’ll see now whether the UK will ask for a referral to the Grand Chamber – it probably will. It should be noted that if the Grand Chamber takes the case, it can deal both with the Article 1 jurisdiction issue and with the merits. I personally don’t think that the Grand Chamber would rule any differently on the merits, but the extraterritorial application issue may be revisited, for better or for worse (probably worse). Most importantly, the Grand Chamber will be holding hearings in the Al-Skeini and Al-Jedda cases on 9 June this year, and they also raise questions of extraterritorial application and norm conflict. Exciting times ahead in the ‘Al-’ cases!

Further commentary will follow.

Dr Lorand Bartels is University Lecturer in Law, University of Cambridge. His publications include Human Rights Conditionality in the EU’s International Agreements (2005, OUP) and Regional Trade Agreements and the WTO (co edited with F. Ortino, 2006, OUP)

The European Court of Justice decided an interesting case last week (Case C-386/08, Brita, 25 February 2010). The Hauptzollamt Hamburg-Hafen (the main customs office of the port of Hamburg) had refused to give preferential treatment under the EC-Israel Association Agreement to products manufactured by an Israeli company in the West Bank.The judgment gives the following facts:

32. The German customs authorities provisionally granted the preferential tariff applied for, but commenced the procedure for subsequent verification. On being questioned by the German customs authorities, the Israeli customs authorities replied that ‘[o]ur verification has proven that the goods in question originate in an area that is under Israeli Customs responsibility. As such, they are originating products pursuant to the [EC-Israel] Association Agreement and are entitled to preferential treatment under that agreement’.

33. By letter of 6 February 2003, the German customs authorities asked the Israeli customs authorities to indicate, by way of supplementary information, whether the goods in question had been manufactured in Israeli-occupied settlements in the West Bank, the Gaza Strip, East Jerusalem or the Golan Heights. That letter remained unanswered.

34. By decision of 25 September 2003, the German Customs authorities therefore refused the preferential treatment that had been granted previously, on the ground that it could not be established conclusively that the imported goods fell within the scope of the EC-Israel Association Agreement. Consequently, a decision was taken to seek post-clearance recovery of customs duties amounting to a total of EUR 19 155.46.

One might have thought that the question would hinge on whether the origin of the products fell within the territorial scope of the EC-Israel Agreement (the ‘territory of the State of Israel’). But the Court took quite a different route. It referred to the EC-PLO Association Agreement, which provides for free trade for products from the ‘territories of the West Bank and the Gaza Strip’ and said:

52. Accordingly, to interpret Article 83 of the EC-Israel Association Agreement as meaning that the Israeli customs authorities enjoy competence in respect of products originating in the West Bank would be tantamount to imposing on the Palestinian customs authorities an obligation to refrain from exercising the competence conferred upon them by virtue of the abovementioned provisions of the EC-PLO Protocol. Such an interpretation, the effect of which would be to create an obligation for a third party without its consent, would thus be contrary to the principle of general international law, ‘pacta tertiis nec nocent nec prosunt’, as consolidated in Article 34 of the Vienna Convention.

Is this correct? Recognizing Israeli competence in relation to products originating in the West Bank does not amount to a denial of Palestinian competence over those products. And even if it did, it does not impose any obligation on the Palestinian authorities not to exercise this competence. They remain free to do so, if they can. So this is not entirely convincing.

The more interesting question is why the Court found it necessary to adopt this odd approach to the case. Why not just determine whether or not the West Bank is part of the ‘territory of the State of Israel’ (as did the A-G)? Could this have anything to do with possible future cases involving annexed territories?

This as hot a case as it gets, and we’ll see what the Chamber does with it. It certainly moved very quickly, since it pronounced on admissibility only in July last year. Whatever the outcome, it is likely that the case will also be referred to the Grand Chamber. Analysis and commentary will follow!

In my view, this will simply not happen. Ever. I might well eventually be proven wrong, of course, but it seems to me that the Falklands dispute is, as a political matter, almost singularly unsuitable for judicial resolution. Here’s why:

First, the current oil exploration dispute cannot judicially be resolved on its own, since it legally entirely depends on who was title over the islands – the UK or Argentina. If it was Argentina who was the Islands’ proper owner, it would be perfectly within its rights to oppose the UK’s implementation of oil exploration by any non-forcible means. If, on the other hand, it was the UK who had title, then it is clear under the UNCLOS and other applicable law that it has every right to drill away, come what may.

Second, as for title, the issue is extremely complicated. To brutally simplify it, Argentina claims title either through succession from Spain, or by having occupied the Islands on its own shortly after gaining independence. The UK relies on prior discovery, effective occupation since 1833, and prescription. It also relies on the Islanders’ right to self-determination, which they’ve freely exercised by choosing to remain a part of the UK. This is, for example, how the UK’s Ambassador to the UN has just stated the UK’s position:

As British Ministers have made clear, the UK has no doubt about its sovereignty over the Falkland Islands, South Georgia and the Sandwich Islands. This position is underpinned by the principle of self-determination as set out in the UN Charter. We are also clear that the Falkland Islands Government is entitled to develop a hydrocarbons industry within its waters, and we support this legitimate business in Falklands’ territory.

Third, to be blunt, the British statement that they have ‘no doubt’ about their title over the Falklands is total rubbish. Privately (of course) they have every reason to doubt it. In fact, I think it would be fair to say that despite the UK’s de facto control for all these years, it is indeed Argentina that has a somewhat superior title over the Islands. Likewise, the Islanders’ claim to self-determination is dubious for various reasons, and UN practice with regard to the Falklands does not support it. For reasons of space and time I will not venture into this further, but there are two recent exhaustive treatments of the subject which are helpful: R. Laver, The Falklands/Malvinas Case (Nijhoff, 2001); R. Dolzer, The territorial status of the Falkland Islands (Malvinas): past and present (Oceana, 1993).

Fourth, following from three above, the UK knows full well not only that there would be a chance, but that there would be a good chance that it might lose a judicial dispute over the Falklands.

Fifth, the UK has invested an enormous amount of political capital in preserving its sovereignty claim over the Falklands, both internally and externally. It has fought a war over them, which still has a place in the national psyche. It has guaranteed to the population (if perhaps not the ‘people’) of the Falklands the right to determine their own fate. For the foreseeable future, it is politically inconceivable that the UK would be willing to renounce this claim, which it would have to be prepared to do if it submits the case to judicial resolution. Not to mention the fact that an oil bonanza would only render such an option less likely.

Sixth, as a matter of fact, the UK’s hold over the Falklands is strong. It’s military position today is far superior to what it was back in the day when the Argentine junta decided on its little adventure. Argentina has no practical way of forcing the issue.

In sum, because of (1)-(6), it is unlikely in the extreme that the UK would be willing to submit this case to a court. It would of course do so if Argentina would be willing to accept arguendo the UK’s title over the Islands, and thus narrow the dispute down to the current oil exploration issues. Yet Argentina has no interest in doing so, because it also knows that it would lose this dispute if title were out of the picture.

So, the only way forward are negotiations. Such negotiations could probably only be successful if title was kept out of the picture, in exchange for a deal on oil rights and a share of profits. The UK and Argentina had such an agreement in 1995, but Argentina repudiated it in 2007. Whether a new deal on those lines is possible today depends on various political considerations that I know nothing about. I am convinced, however, that little else is practically possible.

Anyway, those desperately wanting to see the Falklands dispute (or a simulacrum thereof) argued in court may wish to come to Washington, DC, from 20-27 March, for the international rounds of this year’s Jessup moot court competition…

I. Partial Confirmation of Hay

In its decision, HM Treasury v Mohammed Jabar Ahmed and ors (FC); HM Treasury v Mohammed al-Ghabra (FC); R (on the application of Hani El Sayed Sabaei Youssef) v HM Treasury [2010] UKSC 2, the UK Supreme Court largely confirms the High Court’s approach in Hay, and quashes in part the UK’s ‘Al Qaida Order’ (‘AQO’) because it removes the right of access to an effective remedy (see paras 81-82). The AQO is the implementing measure adopted by the UK Executive to give effect to 1267 sanctions. It is subject to the UN Act 1946, which the Court found not to allow the Executive to remove individual rights. The Court also reverses the decision of the Court of Appeal in A, K, M, Q & G, quashing in part the ‘Terrorism Order’, adopted to implement the 1373 regime. The Law Lords clearly distinguished between the two sanctions regimes, one imposing ‘strict’ obligations, and the other allowing for a margin of appreciation (see paras 64, 148, 196 seq and cf the CFI in OMPI at paras 100-102). What is particularly important in the Supreme Court’s decision is that most of the Law Lords fully accept that the domestic implementing measure of the 1267 regime, the AQO, is strictly conditioned by the relevant Security Council Resolutions. The Court clearly finds that subjecting implementation measures to parliamentary scrutiny could lead to the UK breaching its international obligations under the Charter if the implementing measure was defeated in Parliament (paras 47-49). Lord Brown, dissenting, implies that the Court, in quashing the AQO, would force the UK to flagrantly violate the UN Charter (para 204).

II. The Article 103 UN Charter Issue

Still, even though (implicitly) accepting that they would be forcing the UK to breach its international obligations, the Law Lords quashed in part the AQO. They were able to avoid the overriding effect of Article 103 of the UN Charter, confirmed by the House of Lords in Al-Jedda, by clearly stating that the rights sought to be protected by the applicants in the instant case were invoked on the basis of UK domestic, rather than international, law . It can be said that counsel for the applicants had clearly adopted from the beginning a strategy of basing their arguments on domestic law only, in order to anticipate the Treasury’s reliance on Art. 103. In fact this is clear both in counsel’s argument in Hay before the High Court, and in the fact that the only counsel that argued (in the alternative) under the ECHR before the Supreme Court flat-out conceded that Al-Jedda was against him but invited the Court to reconsider! (paras 66 seq). The Court confirmed Al-Jedda, but went on to state that this does not clarify the position with respect to rights enjoyed under domestic law (see para 75). Lord Rodger, at para 174, made it clear that he was concerned with domestic law rights because rights under the ECHR would be caught by Article 103 UNC. It was indeed by relying on the domestic law right of access to a court that the Court finally quashed the AQO.

III. Dualism—Solange—SC Res 1904: Pressure Leads to Concessions

This reasoning of the Court will have many up in arms with respect to its striclty—even radically—dualist approach. However, that ‘radical dualism’ seems in fact to be ill-disguised exasperation of the English courts with the 1267 regime (see eg para 45). Many epithets have been used by English judges to characterize the regime, and none of them are particularly kind or approving. Their most important complaint—like that of many other domestic courts—is the complete lack of access to an effective remedy and the complete lack of any guarantees of judicial protection. Here, much more than about a domestic right, we are talking about a right guaranteed under international treaties and customary law (even if, to avoid Art. 103, the parallel is not drawn by the Court explicitly; still it is implicit in some of the Law Lords’ opinions). In fact, there seems to be a Solange I undercurrent in the Supreme Court’s decision, much as there was one in the ECJ’s Kadi: when stating that the problem with the 1267 regime is the lack of access to a court at a national or international level, the Court could be seen as leaving the door open to future deference to decisions of the Council, should an adequate mechanism for rights protection be established at the UN level.

On 17 December 2009, the Security Council adopted Resolution 1904 (2009), which can be seen as a direct response to the challenges of designations under the 1267 regime in domestic courts, and as an attempt by the Council to anticipate further challenges. After ‘[t]aking note of challenges, both legal and otherwise, to the measures implemented by Member States under [the 1267 regime]’ (at 9th pre-amb), the Council establishes an ‘Office of the Ombudsperson’ (at para 20). In accordance with its mandate, detailed in Annex II of the Resolution, the Office of the Ombudsperson will receive de-listing requests by individuals subject to the 1267 regime and assist the 1267 Committee through information gathering, engaging in dialogue with the interested parties, and presenting a comprehensive report to the Committee, which will be taking the decision on the de-listing request. The Council notes that ‘the Ombudsperson shall perform these tasks in an independent and impartial manner and shall neither seek nor receive instructions from any government’ (at para 20).

The UK Supreme Court rejected this introduction of an Ombudsperson as a radical step towards creating a regime that would deserve its deference. While the move ‘is to be welcomed’ (see para 78), the Law Lords still considered that the 1267 regime does not offer any access to effective judicial remedies (see paras 78, 80, 239). More must obviously be done on the part of the Council to address the courts’ concerns.

In this post, I would like to add a few thoughts on how the indeterminacy of state practice and opinio juris has caused an indeterminacy in the law, which I don’t think can be denied or removed by any legal analysis, no matter how exhaustive and competent.

Consider, first, that we are (I think) in broad agreement that the prohibition on the use of force in international law, as set out in Article 2(4) of the UN Charter, operates exclusively between states. If, instead of from Afghanistan, Al-Qaeda operated from and launched the 9/11 attacks against the United States from Antarctica, the high seas or some Dr. Evil-style terra nullius volcanic rock in the middle of nowhere (or the Moon even), the jus ad bellum would not in any way limit the US response to the armed attack. It is only if in responding to the attack the US has to encroach on the sovereignty of some other state that Article 2(4) is engaged. (See, similarly, our discussion regarding Israel’s intervention in Gaza). Of course, that in such a situation the jus ad bellum would be inapplicable does not mean that the use of force might not be regulated by some other body of law, such as humanitarian law or human rights law – but that is for now beside the point.

Consequently, it is only if Article 2(4) is engaged that the Article 51 right of self-defence comes into play, since it is an exception to the prohibition on the use of inter-state force. Sure, one could generally say that in responding to an attack by a non-state actor which does not operate from the territory of another state the attacked state is acting in self-defence. But such a characterization would be merely descriptive – this would not be the Article 51 notion of self-defence, which is relevant only when the attacked state responds by violating the sovereignty of some other state. This was, I believe, the case with the Israel’s incursion into Gaza.

However, and this is a crucial point, that Article 51 operates as an exception to the ban on the use of inter-state force, does not logically and conceptually entail that the ‘armed attack’ within the meaning of Article 51 must be attributable to a state. Such a conclusion may follow from an examination of state practice, but it simply does not follow from the text of the Charter or from some broader legal principle, nor from the fact that Article 2(4) itself is inter-state in operation.

So, with this in mind let’s turn to the 9/11 attacks as the paradigmatic example of a non-state actor attacking one state while operating from the territory of another state. The international community has been practically unanimous that the US invasion of Afghanistan was a lawful exercise of self-defence, though some have expressed certain doubts as to the proportionality of the regime-toppling intervention. If the US invasion is accepted as lawful, as I think it must be, there are two, and only two, possible explanations for its lawfulness:

(1) Article 51 requires that the ‘armed attack’ be attributable to a state, thereby engaging its responsibility. Therefore, the 9/11 attacks must have been attributable to the state of Afghanistan. However, the general rules of attribution of acts of non-state actors to states, as articulated by the ILC in its Articles on State Responsibility and by the ICJ in the Nicaragua and Genocide cases, do not allow for a reasonable interpretation that would attribute the 9/11 attacks to Afghanistan, because they require proof that Afghanistan either (a) had complete control over Al-Qaeda, rendering it a de facto state organ; or (b) that Afghanistan had effective control over Al-Qaeda’s conduct in question, i.e. the 9/11 attacks. Since there is no proof of either – indeed, far from it – the 9/11 attacks cannot be attributed to Afghanistan under the general rules. But, because the US invasion was a lawful exercise of self-defence (so we all agree), the attacks MUST have been attributable to Afghanistan. Therefore, the general rules of attribution have either changed, or lex specialis rules of attribution have emerged, whether confined to ‘terrorist’ armed attacks or to the jus ad bellum more broadly, to allow for looser standard of attribution, such as harbouring terrorists or complicity in their actions.

(2) Article 51 does NOT require the attribution of the armed attack by a non-state actor to a state. Rather, for the attacked state to respond against the non-state actor which is operating in another state, the conduct of this latter state must be such to justify the ensuing violation of its sovereignty. Various authors have proposed different standards, but three scenarios seem generally possible, on a scale from the most to the least stringent: (a) the territorial state was complicit or was actively supporting the non-state actor in its armed attack; (b) the territorial state failed to exercise due diligence, i.e. it did not do all that it could reasonably have done to prevent the non-state actor from using its territory to mount an armed attack against another state, or is not doing all it can to prevent further attacks; (c) the territorial state may have exercised due diligence, but it was nonetheless unable to prevent the attack, or to prevent further attacks. The US post-9/11 invasion of Afghanistan could be quite comfortably justified under these standards.

One of these two options MUST be true if the US invasion was a lawful exercise of self-defense. No third option exists, as far as I can see. True, the two options may not lead to different ultimate results in the same factual situations. They are, however, conceptually significantly different and have other, broader implications. The first option couples the jus ad bellum and the law of state responsibility, while the second does not. If the first option is true, then the law of state responsibility must change whenever the jus ad bellum needs to accommodate a more lax standard of self-defense in response to armed attacks by non-state actors. And this is precisely what Christian argues in his initial EJIL piece – that a less stringent rule of attribution which is specific to the jus ad bellum context has evolved. Similarly, see Steve Ratner’s article in 96 AJIL 905 (2002).

Personally, I find the second option to be significantly more preferable to the first, because it maintains the distinction between primary and secondary rules that was famously introduced by the ILC project on state responsibility. That distinction serves several purposes. It helps us solidify general rules of state responsibility while avoiding politically sensitive questions, such as initially rules on treatment of aliens, or now self-defense against non-state actors. It relegates the discussion of these politically thorny questions to the body of primary rules that regulate them – here to the jus ad bellum. Thus, for instance, Kimberley has persuasively argued in her ICLQ article that it is the customary necessity condition for the lawfulness of self-defense that governs an attacked state’s response to an armed attack by a non-state actor operating from another state. We may disagree in precisely what circumstances on the (a) – (c) spectrum that I have outlined above such necessity would exist, but this is precisely the area where such debates should be had, instead of tinkering with the law of state responsibility. Finally, there is an inherent value in having general, coherent secondary rules of attribution which apply irrespective of the primary rules that are being applied, as this enhances the systemic character of international law, and helps it resist fragmentationist impulses.

In other words, it is as a general matter undesirable to have rules of attribution which are primary rule-dependent. Of course, that this is undesirable doesn’t mean it can’t happen. Article 55 of the ILC ASR expressly allows for secondary rules which are lex specialis. But the rules of attribution are precisely the core concept of the law of state responsibility, where lex specialis should not be accepted lightly. This is especially so because it is hard to justify in principle, say, why there should be special rules of attribution when it comes to the use of force by non-state actors pure and simple, but no such rules when it comes to how that force is used, e.g. in regard of genocide or crimes against humanity, as in the Bosnian Genocide case.

Be that as it may, my main point is this: even though I strongly prefer the second option, I still have to say that both options are on the table. For the time being at least, it is objectively impossible to tell which option is valid from an examination of state practice, opinio juris, and the case law. The ICJ has expressly reserved its position on whether Article 51 requires attribution of the armed attack to a state in Congo v. Uganda, despite its earlier pronouncements in Nicaragua and in the Wall case (the latter in my view being better explained by the fact that the armed attacks in question emanated from non-state actors operating from Israeli-occupied territory, thereby disengaging Article 2(4), and consequently Article 51). As for state practice, the official pronouncements of states in the relevant cases are far too ambiguous to say one way or the other whether states consider that Article 51 requires attribution, but that lex specialis rules of attribution have emerged, or instead that no attribution requirement exists.

Take for example the 2006 Lebanon war. Israel was attacked by Hezbollah, a non-state actor operating from within Lebanon. Israel responded by invading Lebanon, thereby engaging Article 2(4), and requiring a justification under Article 51. The reaction by states and other actors has generally been that Israel was entitled in principle to act in self-defense, but that its actions were disproportionate (e.g. because it did not exclusively target Hezbollah). But neither the Israeli official position on its use of force nor the international reaction to it were clear enough to conclusively decide between options (1) and (2).

For example, on 12 July 2006 Israeli Prime Minister Ehud Olmert made the following statement:

This morning, actions were carried out against IDF soldiers in the north. At this time, the security forces are operating in Lebanese territory. The cabinet will convene this evening in order to approve the continuation of the activity.

I want to make it clear: This morning’s events were not a terrorist attack, but the action of a sovereign state that attacked Israel for no reason and without provocation. The Lebanese government, of which Hizbullah is a member, is trying to undermine regional stability. Lebanon is responsible and Lebanon will bear the consequences of its actions.

Here we obviously have a claim by Israel that the armed attacks against it from Lebanese territory were attributable to Lebanon as a matter of state responsibility. But only a few days later, on 16 July 2006, after the Lebanese government denied that it was responsible for the attack, the Israeli position was different:

Prime Minister Olmert emphasized that Israel is not fighting Lebanon but the terrorist element there, led by Nasrallah and his cohorts, who have made Lebanon a hostage and created Syrian- and Iranian-sponsored terrorist enclaves of murder.

Here Lebanon was no longer the author of attacks and responsible for them, but the ‘hostage’ of a non-state actor. This to me sounds a lot like option (2), but there is nonetheless no way to say this with certainty. And the reactions by other states are even more ambiguous (see, e.g. the Security Council debates in S.PV/5489), though to me it seems that many states have accepted Israel’s claim to self-defense without saying that Lebanon was responsible for the attack.

In other words, not even the state using force, Israel, actually articulated a clear position either in favour of option (1) or option (2). Nor did for, that matter, the United States when it invaded Afghanistan. And when the states responding in self-defense refrain from articulating a clear position – probably because they believe that there is some benefit in the ambiguity – then the reactions by other states are almost by definition useless in choosing between the two options. Sure, these states might tell us whether they thought that the use of force was lawful, but they would not explain the theory on which they base their conclusion.

To me, therefore, the conceptual questions of self-defense against non-state actors remain utterly indeterminate while state practice remains as it is. We as lawyers may certainly choose between the two options, but we do so solely – solely – on policy grounds. There’s nothing wrong with that, mind you, and all of the reasons I’ve given above favouring option (2) are no more, and no less, than reasons of policy. And if this is so, there’s also nothing wrong in openly acknowledging the law’s indeterminacy.