By Tobias Thienel

The Grand Chamber of the European Court of Human Rights has today rendered judgment in one of the most – technically and morally – difficult cases to have come before it recently. The case is N v United Kingdom, and concerned the expulsion of someone suffering from AIDS at an advanced stage to Uganda. That person, Ms N, was currently undergoing treatment in the UK, but would not foreseeably receive anything like the same degree of medical assistance in Uganda. It was therefore likely that removal from the UK would expose her to serious pain and discomfort from her disease, and hasten her death. The applicant therefore argued that to remove her to Uganda in those circumstances would be in violation of Article 3 of the Convention. That article, of course, prevents the Contracting States of the ECHR from expelling or extraditing anyone to a situation in the receiving State in which the removed person would face a real risk of ill-treatment contrary to Article 3 standards (that’s torture and inhuman and degrading treatment) (Soering v United Kingdom, paras 85-91). There is also good precedent for the application of that prohibition to the case of an AIDS patient being removed to a poor State where the patient would be without any assistance, and would soon die: that case is D v United Kingdom, and today’s judgment was largely about distinguishing or applying that precedent. The Grand Chamber distinguished D, and found no violation of Article 3 in the present case. I find that difficult to support, and badly argued in the judgment. The Court began by reviewing the case-law resulting from D itself and from its progeny. It pointed out that the Court had in no case subsequent to D found a violation in the context of the removal of AIDS sufferers (para 34 of N). [The Commission had done so, as the Court today acknowledges in para 35.] A sequence of cases had distinguished D on the basis that the other applicants had not been terminally ill, and/or probably would receive some degree of family and medical assistance (paras 32-41 of N). That much is obviously correct. The Grand Chamber then turned to ‘The principles to be drawn from the case-law.’ The Court there drew attention to the fact that D had expressly been decided on very exceptional facts. According to today’s Grand Chamber,

In the D. case the very exceptional circumstances were that the applicant was critically ill and appeared to be close to death, could not be guaranteed any nursing or medical care in his country of origin and had no family there willing or able to care for him or provide him with even a basic level of food, shelter or social support.

The Grand Chamber went on to say that

it considers that it should maintain the high threshold set in D. v. the United Kingdom and applied in its subsequent case-law, which it regards as correct in principle, given that in such cases the alleged future harm would emanate not from the intentional acts or omissions of public authorities or non-State bodies, but instead from a naturally occurring illness and the lack of sufficient resources to deal with it in the receiving country.

Now, it is obvious that the threshold test of Article 3 is not only satisfied where a victim of ill-treatment is put in serious danger of death. The infamous Yoo/Bybee memo is quite wrong, as I suppose is reasonably well-known. The European Court certainly would not suggest otherwise. It has said, instead, that the permissibility of expulsion under D would depend on the presence of such extreme facts. It would be illogical to suggest that this is because of the specific context of expulsion or extradition: the Soering dimension of Article 3 obliges a Contracting State of the ECHR not to expose people to treatment abroad that, if it occurred domestically, would breach Article 3 (Soering, cited above, para 91). That is because Article 3, as an absolute guarantee applicable at all times, demands that States should not countenance in any way any violation of its terms (ibid, para 88). Article 3 therefore catches in the expulsion context everything that it would catch in its ordinary domestic application. The difference between the general application of Article 3 and the much higher threshold now derived from D must therefore lie in the application of Article 3 to the facts on the ground. In other words, the Grand Chamber appears to say that, forgetting all about the expulsion context, the same degree of suffering may breach Article 3 if occasioned by the State, and be compatible with the article if the State is not to blame. I find that difficult to follow. The intensity of the suffering, as determined by reference to a number of pointers, decides whether Article 3 applies (Ireland v United Kingdom, paras 162, 167; Soering, cited above, para 100; Pretty v United Kingdom, para 52). If that is so, Article 3 must either apply to a given intensity of suffering, or not. It does not matter who is responsible for such suffering. Indeed, that’s also what the memorable Joint Dissenting Opinion of Judges Tulkens, Bonello and Spielmann in N argues (para 5). But even if it mattered whether the State was at fault for the suffering of the victim, that still could not support the distinction drawn by the Grand Chamber in N. It is the essence of the Soering obligation, already adverted to above, that a Contracting State of the ECHR must not create a danger of treatment contrary to Article 3 standards by sending someone to a foreign State where such a danger exists, beyond the control of the removing State. In other words, a State has a duty to protect from Article 3 harm, to be discharged by failing to send persons to areas posing a risk of such harm. It follows that in cases like D v United Kingdom or N v United Kingdom, there is in fact a State responsible for the applicant’s suffering; that State is not the receiving State, because it may not have been able to alleviate the applicant’s plight. No, the responsible State is the United Kingdom, at fault for having created a real risk of the ill effects of its own action. It is for this reason that the prospect of suffering for which the receiving State is in no way (even hypothetically) responsible can make the sending State responsible for the expulsion: the expelling State certainly cannot be responsible on the basis that it has exposed the applicant to a (hypothetical) violation of Article 3 ECHR by the receiving State, because this could not catch the situation where, as the Grand Chamber has said today, the harm would in no way be attributable to the receiving State. [The Grand Chamber has thus disapproved the analysis of the House of Lords in R (Bagdanavicius) v Secretary of State for the Home Department [2005] UKHL 38, [2005] 2 AC 668, paras 24-26] It matters only, for the responsibility of the expelling State, that treatment contrary to Article 3 standards be possible. The receiving State need not be at fault for such ill-treatment, but the expelling State will be. I therefore find it preferable to say that Article 3 does not distinguish, for present purposes, between harm caused by a State and suffering following from natural circumstances. The very high threshold adopted in N (from D), which would require a near-death situation, is therefore very difficult to defend. The Grand Chamber went on to say this (para 44):

inherent in the whole of the Convention is a search for a fair balance between the demands of the general interest of the community and the requirements of the protection of the individual’s fundamental rights (see Soering v. the United Kingdom, judgment of 7 July 1989, Series A no. 161, § 89).

This is a very surprising invocation of a very unsuccessful passage from one of the Court’s more famous cases. This particular point, as applied to Article 3, has been fairly clearly disapproved of in Chahal v United Kingdom (para 81) and again in Saadi v Italy (para 127). This part of Soering is therefore no longer really good authority; nor is it correct. It is generally true that the Convention does not in a one-sided manner only protect human rights, but its various limitation clauses are designed to accommodate the interests of the State and of a harmonious society. Even so, it is not a coincidence that Article 3 is without limitations; contraventions of the article are always unacceptable (see also my comments on Saadi here). Considerations of proportionality therefore have no place in Article 3, except as regards the State’s positive obligations (see (Limbuela v. Secretary of State for the Home Department [2005] UKHL 66, [2006] 1 AC 396, para 55 – cited by the joint dissent in N). The last reason advanced by the Grand Chamber may have more to it. The Grand Chamber wrote (para 44):

Advances in medical science, together with social and economic differences between countries, entail that the level of treatment available in the Contracting State and the country of origin may vary considerably. While it is necessary, given the fundamental importance of Article 3 in the Convention system, for the Court to retain a degree of flexibility to prevent expulsion in very exceptional cases, Article 3 does not place an obligation on the Contracting State to alleviate such disparities through the provision of free and unlimited health care to all aliens without a right to stay within its jurisdiction. A finding to the contrary would place too great a burden on the Contracting States.

As the dissent notes, this passage betrays the Court’s real concern. The dissent engages the point by showing that the statistics are nowhere near as worrying as the majority of the Grand Chamber may have thought. Nonetheless, the majority does have a point, regrettable though it is. However, as I have just attempted to argue, that concern cannot be brought into the Article 3 enquiry in the form of a proportionality test. Still, it may obtain some relevance by another route: compliance with the obligation not to expel necessarily means that the applicant will stay within the UK, and will remain entitled to healthcare. Compliance with the negative obligation at issue in N would therefore have established the factual basis of a positive obligation. It is, however, in the nature of classic human rights protection that positive obligations are less easily accepted than are the ordinary negative ones. The fact that the negative obligation not to expel would entail a further positive obligation therefore might properly be taken into account in deciding whether the negative obligation applies. However, I’m not at all sure this is so: it might as easily be argued that, if a negative obligation brings with it a subsequent positive obligation, then so be it, precisely because negative obligations are at the heart of the Convention, and cannot be displaced by considerations relating to ancillary, positive obligations. In sum, I believe the Court had before it an unenviable task. I can understand its political desire to dismiss the application. However, I cannot necessarily agree with the decision, and certainly have real trouble with the reasoning. I would suggest the Court made a few very unfortunate statements on its way to the result it felt convenient. Whether these will now hold up as precedent may be another matter…