I am working on a post that dives into the core issue in Mohammed v. Ministry of Defense (MOD)—whether the law of armed conflict (LOAC) permits security-based detentions in non-international conflict and what the relationship is between that legal issue and applicable human rights law. I am encouraged by invocations of my 2009 article in the American Journal of International Law by opposing sides of these debates (here and here). So, it will definitely be time for me to roll up my sleeves.

In the meantime, I thought to bring attention to a separate part of the court’s opinion: its holding that Article 5 of the European Convention on Human Rights prohibits “detention solely for the purpose of interrogation with a view to obtaining valuable intelligence.”

That was a fairly predictable outcome. Article 5 of the European Convention allows states to detain individuals only on the basis of a limited list of grounds. And interrogation (i.e., the individual’s intelligence value) is clearly not one of them. The holding, however, raises more complicated puzzles, such as:

  1. Would the European Convention permit interrogation-based detention if a state entered a derogation under Article 15 of the treaty?
  2. For states not party to the European Convention, would the International Covenant on Civil and Political Rights or customary international human rights law permit interrogation-based detention?
  3. Would the law of armed conflict answer these questions differently (especially if the court had to resolve more directly the lex specialis question)?

I don’t attempt to answer all those questions here, but I do discuss one piece of the puzzle: does the law of armed conflict permit detention solely for the purpose of interrogation with a view to obtaining valuable intelligence? It is of course legal to incidentally interrogate individuals who are detained for other purposes (as long as the interrogation is performed humanely and in accord with due procedures), but that is not the question here.

Before getting into the law, it is important to highlight some of the competing normative concerns. A Ministerial statement to Parliament by the MOD explained the government’s compelling interests operating in Afghanistan:

“[I]n exceptional circumstances, detaining individuals beyond 96 hours can yield vital intelligence that would help protect our forces and the local population – potentially saving lives, particularly when detainees are suspected of holding information on the placement of improvised explosive devices.

Given the ongoing threat faced by our forces and the local Afghan population, this information is critical, and in some cases 96 hours will not be long enough to gain that information from the detainees.  Indeed, many insurgents are aware of the 96 hours policy and simply say nothing for that entire period.”

A competing normative concern is that detention for the purpose of interrogation would potentially justify holding individuals on the sole basis of their intelligence value even if they do not pose a security threat (think: the spouses of suspected terrorists). Indeed, former General Counsel at US Special Operations Command Central Andru Wall warned: “detaining individuals on the basis of what they were believed to know could be a slippery slope leading to mass, unwarranted detentions.”

So what does the law—or LOAC in particular—have to say? In a 2009 article for the US Naval War College’s International Law Studies, I wrote that LOAC does not permit long-term detention for the purpose of gathering intelligence. I have appended below an excerpt from that article. It begins with a description of some US detention policies at the time.

 A remaining question is whether the United States can detain individuals, on a long-term or indefinite basis, for the purpose of gathering intelligence. Before analyzing that question of law, first consider the record of US detention practices following September 11. The government has used intelligence value as a ground for initial internment decisions, as well as for denying release. Former Deputy Assistant Secretary of Defense for Detainee Affairs Professor Matthew Waxman recently wrote: “Intelligence gathering through questioning of those in custody constitutes another important reason for detention in warfare, and especially in fighting terrorist networks.” With respect to the global sphere of operations, the 2006 Counterinsurgency Field Manual states that information gathering provides a reason for detaining two classes of individuals: (1) “persons who have engaged in, or assisted those who engage in, terrorist or insurgent activities” and (2) “persons who have incidentally obtained knowledge regarding insurgent and terrorist activity, but who are not guilty of associating with such groups.” Notably, information gathering appears to be an independent basis for detaining the first category of individuals even if they no longer pose a security threat. However, for the second category, the Counterinsurgency manual states: “Since persons in the second category have not engaged in criminal or insurgent activities, they must be released, even if they refuse to provide information.” It stands to reason that individuals in the first category could be denied release if they refuse to provide information. […]

With respect to detention in Guantanamo specifically, in determining whether a detainee should be transferred to the base, US military screening teams and the combatant commander must consider “the possible intelligence that may be gained from the detainee.” And administrative review boards may consider whether a detainee “is of continuing intelligence value” in deciding whether to recommend release. That standard appears to regularize practices that predated the ARB process. Although stated in a summary fashion, a joint report by UN human rights officials concerning Guantanamo concludes “that the objective of the ongoing detention is not primarily to prevent combatants from taking up arms against the United States again, but to obtain information and gather intelligence on the Al-Qaeda network.”

Within the United States, the cases of individuals such as José Padilla and Ali Saleh Kahlah al-Marri suggest that intelligence value may constitute a dominant rationale for detention. […] In al-Marri’s case, federal judges expressed concern over the apparent interrogation– based reasons for transferring the petitioner from criminal jurisdiction to military administrative detention:

“[N]ot only has the Government offered no other explanation [than interrogation purposes] for abandoning al-Marri’s prosecution, it has even propounded an affidavit in support of al-Marri’s continued military detention, stating that he ‘possesses information of high intelligence value.’ See Rapp. Declaration. Moreover, former Attorney General John Ashcroft has explained that the Government decided to declare al-Marri an enemy combatant only after he became a ‘hard case’ by ‘reject[ing] numerous offers to improve his lot by . . . providing information.’ John Ashcroft, Never Again: Securing America and Restoring Justice 168-69 (2006).


It is important to recognize that intelligence value has also constituted an independent basis for administrative detention in Iraq. Consider Lieutenant Andru Wall’s account of detainee operations:

Officially, individuals could be detained for their intelligence value for no more than 72 hours; however, anecdotal evidence suggested that longer intelligence detentions were common. The argument in favor of intelligence detentions was that, for example, if an individual knew who was responsible for carrying out attacks on Coalition Forces . . . then withholding [this information] constituted an imperative threat to the security of Coalition Forces . . . . The argument against such detentions was that the individual himself did not pose an imperative security threat . . .


Three arguments might be raised to support the legality of US practice. First, the Geneva Conventions contain no express prohibition on the use of detention for intelligence-gathering purposes. Second, detention is permitted if obtaining the relevant information serves an imperative security interest. Third, if a State has the authority to detain an individual until the cessation of hostilities, the State has the prerogative to release her earlier if she provides valuable intelligence information.

At the outset in addressing these arguments we should note that an express provision of the Geneva Conventions may not be necessary if the regime implicitly contemplates that the only basis for detention is to prevent individuals returning to the fight. A customary norm may also suffice if treaties do not. And, even if LOAC permits interrogation incidental to detention, it does not necessarily permit detention for the purpose of interrogation. Nor does it permit coercive interrogation.

Let’s turn to an elaboration of some of these points and other points as well. First, all three arguments are contradicted by legal authorities that have addressed the subject with respect to the general LOAC regime. The ICRC publicly criticized the use of Guantanamo for interrogation purposes. The joint report of UN officials declared: “The indefinite detention of prisoners of war and civilian internees for purposes of continued interrogation is inconsistent with the provisions of the Geneva Conventions.” And a plurality of the US Supreme Court stated in dicta: “Certainly, we agree that indefinite detention for the purpose of interrogation is not authorized.” Some commentators have suggested that the plurality’s statement is conclusory and without citation to legal authority. However, in earlier passages, the opinion references authorities suggesting that detention is permitted exclusively to prevent individuals returning to the battlefield. US policy, accordingly, thwarts the collective judgment of the US Supreme Court (in a plurality opinion), the ICRC and UN human rights officials.

Second, other provisions of the Geneva Conventions indirectly support the conclusion that indefinite or long-term detention is permitted only to prevent individuals returning to the battlefield. In general, detaining powers argue against early release of prisoners of war on the ground that the individuals might return to the fight. However, some detainees are too sick or wounded to return to the battlefield. A valuable question for our purposes is whether the detaining power could nevertheless hold the individual to gather intelligence. The Prisoner of War (POW) Convention is clear: it places a categorical obligation to repatriate such individuals to their home countries. There is no exception for detaining or precluding release of individuals on any other grounds such as intelligence value.

Third, the most relevant rules may not be found directly in provisions regulating detention. The most relevant source may be found in rules governing interrogation. And those interrogation rules preclude the initial decision to detain an individual, as well as the purported prerogative to order release of a detainee who provides information. More specifically, the use of intelligence value violates Article 17 of the POW Convention and Article 31 of the Civilians Convention. Both articles strictly prohibit physical and moral coercion to obtain information from detainees. Accordingly, individuals who are interrogated should not receive better treatment (release from detention) or worse treatment (continued confinement) on the basis of whether they provide or withhold information. In short, the relevant LOAC rules are found more directly in provisions regulating methods of interrogations, rather than provisions regulating grounds for detention. Notably, the former constitutes an independent basis for the application of LOAC in noninternational armed conflicts. That is, even if LOAC does not regulate unlawful confinement in non-international armed conflict, it undoubtedly regulates coercive interrogations.

Fourth, an individual’s possession of information does not constitute a valid security rationale for internment under the Civilians Convention. According to the ICRC Commentary, States have significant discretion to define activities that threaten their security. The Commentary, however, also suggests that the individuals must themselves directly pose the threat. The paradigmatic examples provided by the Commentary include “[s]ubversive activity carried on inside the territory of a Party to the conflict or actions which are of direct assistance to an enemy Power.” More specific examples include “members of organizations whose object is to cause disturbances, or . . . [individuals who] may seriously prejudice its security by other means, such as sabotage or espionage.” […]

Finally, the implications of allowing intelligence value as an independent ground for long-term or indefinite detention are intolerable. Doing so might permit the confinement of individuals, such as the children or other family members of combatants, who have no engagement in hostilities but have personal knowledge about the combatants. It might also permit the confinement of innocent detainees who do not have information themselves but are held as bargaining chips to coerce other individuals to provide information. […]