Repatriating the Children of Foreign Terrorist Fighters and the Extraterritorial Application of Human Rights

UPDATE: The CRC admissibility decision referred to in the post is no longer publicly available via ODS. I am unsure as to why that’s the case. Readers can find it here.

Now that we have all (sort of) recovered from the US elections, back to our regular programming: last week the UN Committee on the Rights of the Child published its views in a major case (L.H. et al v. France, communications No.79/2019 and No. 109/2019, CRC/C/85/D/79/2019–CRC/C/85/D/109/2019), finding that France (and, accordingly, other similarly situated states) had jurisdiction over children in Kurdish-controlled camps in Syria who are French nationals, and are there because of their parents’ involvement with ISIS. The upshot of the (unanimous) decision is that states of nationality have the positive obligation to protect the human rights of child nationals in the Syrian camps, despite the fact that these camps are under the control of a non-state armed group. And while the decision was nominally on the jurisdictional issue only, it is inevitable that the Committee will find on the merits that France has the duty to repatriate its child nationals from the camps. In so doing, the Committee took the same basic position as a recent paper on the topic by the UN Special  Rapporteur  on  the  promotion  and  protection  of  human  rights  while  countering terrorism and the UN Special Rapporteur on extrajudicial, summary or arbitrary executions.

In this post, I will analyse the extraterritoriality holdings of the Committee and the two Rapporteurs. They both embrace an expansive approach to extraterritoriality, and reach the most human rights-friendly of outcomes – normally very much a good thing. But the reliance on nationality – as we will see, somewhat hedged and cocooned in a supposedly functional analysis – is to my mind deeply problematic and has great potential for arbitrariness. This is one of those cases, in other words, in which the apparently morally right outcome may have a very dark undercurrent. Let me explain how and why.

As a preliminary matter, under the two established tests of extraterritorial jurisdiction under human rights treaties, the children in the Kurdish-controlled camps would clearly not be within the jurisdiction of France and other states of nationality. First, under the spatial model of jurisdiction, conceived of as effective overall control over territory, the camps were not under French control. Second, under the personal model of jurisdiction, conceptualized as authority or control over an individual, again the children and their guardians were in Kurdish custody and at least physically not under French control – France did not actively do anything to them, but simply refused to repatriate them.

That said, two further points need to be borne in mind. There are references in the extraterritoriality jurisprudence of human rights bodies to an idea that decisions made on a state’s territory that produce effects outside the state’s territory trigger jurisdiction in the sense of human rights treaties (cf. para. 9.6 of the Committee’s views). But not only has this effects model never really consolidated in the case law – and in fact it both lacks any internal coherence and any limiting ability for a threshold criterion – it is completely non-sensical to use it with regard to decisions to admit, or not, an individual to a state’s territory. An obligation to protect human rights cannot be triggered by a decision not to protect them; the former is logically prior to the latter. (As a useful comparison point, see this post by Vladislava Stoyanova on the Strasbourg M.N. case and decisions on visa refusal). It would be wholly arbitrary to argue, for example, that a French child whose application for repatriation was formally denied by French authorities was for that reason owed a duty of protection by France, whereas a child in the same position but had never made such an application was owed no duty of protection. Finally, and crucially, the notion of jurisdiction in human rights treaties is not the concept of jurisdiction to prescribe in public international law, which recognizes active and passive nationality as heads of jurisdiction. The former is about an entitlement to human rights vis-à-vis a state, the latter is about the state’s right to regulate certain situations by means of its domestic law (for an extensive discussion, see here). In short, not every British national abroad is within the UK’s jurisdiction for the purposes of the ECHR, the ICCPR or the CRC by virtue of their nationality, even though some parts of British law may continue to apply to them wherever they might be.

Bearing all this in mind, let us now look at what the Committee and the Rapporteurs actually said. The Committee’s reasoning is rather sparse for an issue of this importance – really a single paragraph, which I will quote in full (para. 9.7):

In the present case, the Committee notes that it is uncontested that the State party was informed by the authors of the situation of extreme vulnerability of the children, who were detained in refugee camps in a conflict zone. Detention conditions have been internationally reported as deplorable and have been brought to the attention of the State party’s authorities through the various complaints filed by the authors at the national level. The detention conditions pose an imminent risk of irreparable harm to the children’s lives, their physical and mental integrity and their development. The Committee recognizes that the effective control over the camps was held by a non-State actor that had made it publicly known that it did not have the means or the will to care for the children and women detained in the camps and that it expected the detainees’ countries of nationality to repatriate them. The Committee also notes that the Independent International Commission of Inquiry on the Syrian Arab Republic has recommended that countries of origin of foreign fighters take immediate steps towards repatriating such children as soon as possible. In the circumstances of the present case, the Committee observes that the State party, as the State of the children’s nationality, has the capability and the power to protect the rights of the children in question by taking action to repatriate them or provide other consular responses. These circumstances include the State party’s rapport with the Kurdish authorities, the latter’s willingness to cooperate and the fact that the State party has already repatriated at least 17 French children from the camps in Syrian Kurdistan since March 2019. (emphasis added)

Note that it is unclear what test precisely is being applied by the Committee, except that it’s not one of control over the camps. It also seems reasonably clear that the Committee is not holding that nationality alone suffices to trigger jurisdiction under Art. 2(1) CRC, i.e. it’s not saying that every child of French nationality abroad is within France’s jurisdiction. Perhaps the best reading of the Committee’s reasoning is that it is adopting a functional approach, i.e. that it is holding that France has the duty to protect the children because it has the ability to do so, on the basis of numerous contextual factors, one being the children’s nationality, and another the Kurdish authorities’ willingness to cooperate and release the children to France. However, note also the perverse consequence of the last factor that the Committee took into account – had France not repatriated other children before, this would have presumably worked in its favour in denying that it had the obligation to repatriate the applicants.

This brings us to the Rapporteurs’ paper, which is much more extensive in its analysis (although a lot of that analysis vacillates between several different conceptions of state jurisdiction). The bottom line, however, is strikingly similar to the Committee’s (paras. 35 & 36):

The Special Rapporteurs point out that States are in the best position to ensure the protection of human rights for children and their guardians in camps in the northern Syrian Arab Republic. In the absence of their engagement and acceptance of legal responsibility, children face death, starvation, and extreme physical and emotional harm, as do their mothers. In this context, they note that in the very specific circumstances of these camps in the northern Syrian Arab Republic it is undeniable that the State of nationality for citizens have the only tenable legal claim to protect their citizens, and the capacity to make such claims materialize. The Special Rapporteurs also underscore that the relevant Kurdish authorities have made consistently clear their willingness and capacity to support returns to States and their inability to manage the humanitarian catastrophe they face, a fact that is demonstrated by multiple successful return processes.

States that have de facto control over the human rights of children and their guardians in camps in the northern Syrian Arab Republic have positive obligations to prevent violations of those rights. Whether a State has such control is a question of fact. Relevant factors are likely to include the proximity between the acts of the State and the alleged violation, the degree and extent of cooperation, engagement and communications with the authorities detaining children and their guardians, the extent to which the home State is able to put an end to the violation of the individual’s rights by exercising or refusing any positive interventions to protect and promote the rights of their nationals, and the extent to which another State or non-state actor has control over the rights. (emphasis added)

The Rapporteurs clearly do apply a functional approach – note the language of control over the children’s rights and the states’ of nationality ability to protect them, coupled with a multi-factor contextual test. This conclusion is also buttressed by references in the paper (paras. 14-7) to the Human Rights Committee’s General Comment No. 36, which clearly embraced a functional model. In any event the Rapporteurs’ reasoning is clearer and more rigorous than the Committee’s.

Before I proceed further, I do need to clarify that am very happy with the idea of repatriating children (and not just children) from the squalor of the camps in Syria. But I am still uncomfortable with the reasoning of both the Committee and the two Rapporteurs. Why? Because of that reasoning’s potential for wholly arbitrary line-drawing, as I will now explain.

The main benefit of a functional approach to extraterritoriality – especially in the context of positive obligations – is precisely in the avoidance of arbitrary lines. States would owe duties of protection to individuals when they are in fact capable of providing such protection. The functional model’s drawback, however, is in its inability to serve as any kind of threshold criterion – unless the capacity inquiry is applied with great stringency, it becomes impossible to persuasively argue that one category of people outside the state’s borders deserve protection, whereas others do not. In particular, the protective obligation can become very burdensome indeed for states. This is why in my book I argued for a non-perfect solution, tying positive obligations to territorial control while applying negative obligations without restrictions, a solution that again would not be perfect, but would also not be the enemy of the good.

Which brings me to the arbitrariness potential with regard to repatriation – the central role that the nationality of the children plays in the reasoning of the Committee and the Rapporteurs. Note, in that regard, how very careful they were in not saying that nationality alone is a sufficient basis for the protective obligation to arise. But nationality was nonetheless a ‘but for’ condition in their reasoning – if the children were not French, France would not have had the duty to protect them. And this is troubling, for a number of connected reasons. First, because nationality is generally an accident of birth, and is a poor normative foundation for non-political rights. Second, because the factual premise of the analysis – that the state of nationality is somehow uniquely placed and able to help the children – is demonstrably false. Nothing whatsoever would prevent France from taking in children (or adults) from the camps who are not French nationals – the Kurds would be equally happy to see them go. The position would have been different if the Kurdish authorities said that they were only willing to repatriate individuals to their state of nationality, but to my knowledge they have not said so. Third, because nationality is really being used here as a burden-sharing device to fix a collective action problem, but the purely political nature of that problem is not compatible with the universality of human rights. Fourth, and finally, because of the enormous amount of variation between state laws with regard to the acquisition of nationality, as well as the deprivation of nationality.

Let me develop these points by way of some examples. Imagine a family in one of the camps, of a mother with two children. The mother is a Syrian national. The children have different fathers, both ISIS fighters – one was French (and later died in combat), the other British. One child accordingly has French nationality, and the other British. Are we really going to say that, under human rights law, France has the duty to protect only the French child, but not their British sibling or the Syrian mother? Or that France and the UK have the duty to repatriate the children, each to their own country, and without the mother? Or, consider the family’s immediate neighbour in the camp, a Syrian child, who is gravely ill and desperately in need of medical treatment in the West. Are we really going to say that, under human rights law, that child is less deserving of protection from France than a perfectly healthy child who just happens to be French? Or, consider a pregnant British woman in the camp, whose nationality is stripped away (let’s stipulate lawfully, for the sake of the argument) by the Home Secretary, so that she is unable to transmit British nationality de jure sanguinis to her unborn child. Are we really going to say that, under human rights law, the woman’s prior British children in the camp are deserving of protection by the UK, while her (presumably) Syrian baby is not?

And so forth. I hope to have demonstrated that the potential for arbitrariness inherent in the reasoning of the Committee and the two Rapporteurs is obvious, despite being motivated by the purest of intentions. Such potential becomes even more obvious when we move from this particular context to others in which certain human rights may be conditioned on citizenship (e.g. privacy with regard to extraterritorial surveillance – see here and here, 97-101). And only add to this the sheer uncertainty of how this variant of the functional approach would translate to other situations – e.g. the assistance by states to their nationals in foreign prisons, the provision of medical assistance (say Covid-19 vaccines and therapeutics), etc.

In sum, the central role that nationality plays in the legal analysis of the Committee and the Rapporteurs is highly questionable, as is its role in the repatriation discourse more broadly. That said, I do hope that these children do get the help they deserve as a result of the efforts of the Committee and the two Rapporteurs. I also hope, however, that their reasoning is not uncritically emulated in the future. It sounds so obviously correct to say that the children in the camps need outside help – and I agree – but grounding that claim to help in nationality can lead us down a very dangerous path.