Yesterday’s post provided a brief guide to the UN Human Rights Committee’s General Comment No. 35 (GC) on Article 9 of the International Covenant on Civil and Political Rights (ICCPR), the right to liberty and security of the person. This post analyses the national security aspects of the GC. 

Applicable Law

As I noted in my earlier post:

–       the relationship between IHRL and IHL is currently a pivotal international law issue;

–       deprivation of liberty is probably the area which most obviously exposes the tensions in that relationship;

–       the applicable law for deprivations of liberty arising from the UK’s military operations overseas has been, and continues to be, actively litigated (see Al-Jedda v UK hereHassan v UK here and Serdar Mohammed v Ministry of Defence here) and

–       given the range of views on the matter, it is not an issue which is likely to be resolved soon.

The GC addresses applicable law in paragraph 64 whilst paragraphs 65-66 consider derogation and IHL:

–       The first two sentences of paragraph 64 rightly acknowledge that IHL does not displace or replace IHRL/Article 9 in situations of armed conflict. The Grand Chamber, in Hassan v UK, has recently upheld the same point in the specific context of the ECHR: see my earlier posts on Hassanhere and here.

–       The second sentence makes the additional point that IHL may be relevant for the purposes of the interpretation of Article 9.

–       The third sentence of paragraph 64 (which was inserted after the first reading in March 2014) then states, “Security detention authorized and regulated by and complying with [IHL] in principle is not arbitrary.” The meaning of this sentence is not clear for the following key reasons.

  •    First, what is the relationship between the second and third sentences of paragraph 64? Does the third sentence define the “interpretative role” statement in the second sentence or is it simply one example of it?
  •    Secondly, as the GC itself notes, the second and third sentences of Article 9(1), which address the non-arbitrariness requirement and the lawfulness requirement, “overlap” but remain distinct (see GC, paragraphs 11-12). The third sentence of paragraph 64 considers arbitrariness but there is no consideration of whether detention which is “authorized” by IHL could be capable of satisfying the lawfulness requirement in Article 9(1), viz. “No one shall be deprived of his liberty except on such grounds and in accordance with such procedure as are established by law.”
  •    Thirdly, there is no explanation of what aspect of IHL “in principle” (which is a strikingly material qualification) renders the security detention “not arbitrary”: is it the provision of grounds for detention, the existence of procedural safeguards, a combination of both?
  •    Fourthly, and perhaps most unhelpfully, the third sentence makes no distinction between IHL as it applies in international armed conflicts (“IAC”) and as it applies in non-international armed conflicts (“NIAC”). That is a critical distinction given the ongoing debate and litigation over whether, in a NIAC, IHL “authorizes and regulates” “security detention.” Given its recognition of the link between IHRL and IHL and the interpretative role of the latter (in paragraph 64, second sentence), the GC could, and should, have addressed directly, and in paragraph 64, the question of whether IHL as applicable in NIAC “authorizes and regulates” security detention. The fact that it did not do so is likely to invite different readings of the GC and paragraph 64 in particular. For example: an important clue to the meaning of the third sentence of paragraph 64 is to be found in paragraph 66 which, in the context of derogation, provides, “During international armed conflict, substantive and procedural rules of international humanitarian law remain applicable and limit the ability to derogate, thereby helping to mitigate the risk of arbitrary detention [fn referring to para 3 of General Comment No. 29]. Outside that context, the requirements of strict necessity and proportionality constrain any derogating measures involving security detention, which must be limited in duration and accompanied by procedures to prevent arbitrary application, as explained in paragraph 15 above, including review by a court within the meaning of paragraph 45 above.” (See, similarly, the note of the Committee’s discussions on 22 July 2014, here). This is an interesting passage not only (a) because there is no reference to NIAC (cf. para 3 of General Comment No. 29 which refers to both IAC and NIAC) but, strikingly, (b) because the GC identifies IAC/the applicable IHL as being an exceptional category (“outside that context”). This strongly implies that the reference to “IHL” in the third sentence in paragraph 64 is a reference only to the IHL applicable in IAC. Paragraph 66 is also significant in its own right, and not just for the light it sheds on the meaning of paragraph 64. The fact that the Committee chose expressly to refer to IAC (and not, e.g., to armed conflict in general) in paragraph 66 appears to indicate that it does not regard IHL in NIAC as including rules regarding arbitrary detention – and, therefore, that the only source of protection against arbitrary detention in NIAC is IHRL.


The GC addresses jurisdiction in paragraph 63. The sentence which appeared in the first draft (and which, in my post on that first draft, I suggested should be elaborated) appears, unchanged, in paragraph 63. Whilst the meaning of the sentence is clear it is a very brief reference to an issue which has been, and continues to be, of huge importance in the national security context – as is vividly illustrated by the recent developments regarding the US government’s legal position on the Convention Against Torture, considered in the posts here and here. Nor is such a brief reference easy to reconcile with the fact that, during its July session, “The Committee discussed at length inclusion of deprivation of personal liberty by the action of a State within another State’s territory.”


In my post on the first reading draft of the GC, I commented that:

–       Although the GC considered the grounds on which a person might be deprived of liberty it did not squarely address security detention.

–       Security detention should be more fully considered, especially since the protections of Article 9 are softer than other IHRL protections regarding the right to liberty. For example, a key difference between Article 9 and Article 5 of the ECHR is that the former “does not provide an enumeration of the permissible reasons for depriving a person of liberty” (see GC, paragraph 14). Article 5 of the ECHR does include such a list, e.g. Article 5(1)(c) provides for the following permissible ground for deprivation of liberty: “the lawful arrest or detention of a person effected for the purpose of bringing him before the competent legal authority on reasonable suspicion of having committed an offence or when it is reasonably considered necessary to prevent his committing an offence or fleeing after having done so.”

The GC addresses security detention in paragraph 15:

–       As noted above, this paragraph was inserted after the first reading draft. In this respect, at least, the GC is an improvement on the first reading draft because security detention is now directly considered.

–       Paragraph 15 appears to have given the Committee cause for concern: it was adopted only after the October session (cf. the remainder of paragraphs 1-41 which were adopted after the July session whilst paragraph 15 was only provisionally adopted). Further, the note of the Committee’s discussions on 22 July 2004 records that they “looked into draft paragraph 15 in depth.” Unfortunately, the product of these deliberations is not a happy one – paragraph 15 is opaque.

–       For example, it is not clear whether, or when, the paragraph is intending to address (a) the lawfulness requirement (the third sentence of Article 9(1)) and (b) the non-arbitrariness requirement (the second sentence of Article 9(1)). The location of paragraph 15 in the scheme of the GC does not shed light on its content: it appears after the introductory paragraphs on arbitrary detention and lawful detention, i.e. paragraphs 10-14 and before paragraph 22 which expressly considers the lawfulness requirement. So it may be intended to deal only with arbitrariness/security detention but that is not clear.

–       As to (a): the first sentence of paragraph 15 appears to assert that security detention could be lawful even where “no prosecution on a criminal charge” is in “contemplation.” The third sentence of paragraph 15 then appears to impose three pre-conditions regarding the lawfulness of such security detention: (i) that it occurs “under the most exceptional circumstances,” (ii) that the person detained poses “a present, direct and imperative threat” and (iii) the State has to show that the individual poses such a threat. These pre-conditions (and the reference to “internment” in the first sentence of paragraph 15) are reminiscent of IHL in the context of security detention in IAC but the GC does not identify “the most exceptional circumstances” as being IAC. Nor does the GC cite any authority which might illuminate the provenance or scope of these pre-conditions.

–       As to (b): the second sentence of paragraph 15 appears to assert that even a lawful security detention which is “not in contemplation of prosecution on a criminal charge” will “normally” be arbitrary given the availability of other effective measures of addressing the threat. The only way to overcome this presumption of arbitrariness appears to be for the State to show, among other things, that the threat posed by the individual “cannot” be addressed by alternative measures (see third sentence, paragraph 15). Again, however, there is no cited authority from which further assistance on these issues could be obtained.

Applicable law, jurisdiction and security detention are difficult and politically sensitive issues. It is laudable that the Committee has addressed them in the GC and its reticence may be deliberate e.g. because it could not reach consensus or because it preferred to let these issues develop through other avenues such as concluding observations, the complaints mechanism under the First Optional Protocol and cases before national and regional courts.  Nevertheless, the fact remains that the GC is, in the respects considered above, opaque. It is regrettable that the Committee did not fully seize the opportunity with which it was presented to give clearer “authoritative guidance” on these issues, to the extent that it felt able to consider them.