2.1. Definition of NSAG

Despite the existence of a significant number of NSAGs engaged in NIACs, which represent the predominant type of armed conflict nowadays, conventional IHL does not contain a clear definition of what constitutes an ‘organised armed group’.Footnote ¹³

The idea to set a list of definitive criteria for the establishment of the existence of a NIAC within the meaning of CA3, and thus distinguishing the latter from a ‘handful of individuals [raising] rebellion against the State and attack[ing] a police station’, received attention during the process of adoption of the GCs.Footnote ¹⁴ Although this idea was eventually dropped, the Commentaries on the GCs included a list of non-obligatory ‘convenient criteria’ that were discussed during a phase of negotiations and could potentially evidence the existence of a NIAC.Footnote ¹⁵

Remarkably, one of those criteria pointed specifically to the need for a certain level of organisation on the side of the NSAG to be qualified as a party to a NIAC.Footnote ¹⁶ It is one of the criteria for the determination of the existence of a NIAC that were subsequently confirmed through the practice of international criminal justice institutionsFootnote ¹⁷ and universally accepted as a prerequisite for the application of CA3 and AP II.Footnote ¹⁸

Therefore, the starting point for defining NSAGs is the possession of ‘some degree of organization’.Footnote ¹⁹ While elaborating a list of factors that could be used to indicate a sufficient level of organisationFootnote ²⁰ required by CA3 and AP II, tribunals have repeatedly refused to provide a unified definition or a mandatory checklist of criteria to determine the status of a party to a NIAC, reiterating that in each concrete situation determination has to be made on a case-by-case basis.Footnote ²¹

In reality, NSAGs can choose different modes of organisation, starting from centralised groups establishing military formation resembling state armed forces and creating an elaborate system of judicial institutions, to groups embedded in local communities and lacking any hierarchical structure or permanent leadership.Footnote ²² The development of a uniform legal definition, apart from being a difficult task, would inevitably entail the risk of being too restrictive. At the end of the day some NSAGs, despite being parties to NIAC, would risk dropping out of the scope of application of IHL because of a different organisational structure that was not envisaged by the definition or would be denied such status by states that are interested in labelling the group as a criminal gang rather than a group in possession of even limited international personality.

Thus, in accordance with the following criteria developed in jurisprudence and doctrine, in order for the NSAG to be considered a party to an existing NIAC it has to possess a certain degree of organisation sufficient for:

• • being able to plan and carry out military operations;Footnote ²³ and

• • establishing responsible command allowing for maintaining discipline and ensuring compliance with basic humanitarian obligations.Footnote ²⁴

The same criteria were endorsed in the updated Commentary on GC I, referring particularly to the criterion of organisation under CA3.Footnote ²⁵

When it comes to NSAG parties to NIAC under AP II, apart from an additional requirement of territorial control, reasonably enough both criteria would imply a higher level of organisation providing for:

• • the ability to carry out ‘sustained and concerted military operations’; and

• • the existence of responsible command capable of enforcing obligations under AP II.Footnote ²⁶

After establishing the essential features of the NSAG it is worth mentioning briefly criteria that are not relevant for the purposes of this article: the motivation of the NSAG and its capacity to establish control over part of the territory.

The suggestion to include a specific motivation – for instance, of a political nature – as one of the determinative criteria was discussed by states during the adoption of the GCs and rejected.Footnote ²⁷ This idea has also been criticised in the legal scholarship.Footnote ²⁸ Indeed, neither CA3 nor AP II contains any provision that could be interpreted as differentiating the scope of powers and responsibilities based on the motivation of parties to a NIAC. Therefore, the power to detain has to be studied in respect of all NSAGs regardless of their specific objectives.

Establishing control over a part of the territory was introduced into AP II as a prerequisite for its application, thus limiting the material scope of application to armed conflicts of a certain level of intensity.Footnote ²⁹

At the same time a deliberate decision was taken: to keep AP II separate from CA3, thus maintaining the autonomous existence and broad scope of application of the latter.Footnote ³⁰ It was the intention of the drafters of CA3 to provide albeit limited scope of protection to the maximum possible number of persons affected by adverse consequences of an armed conflict.Footnote ³¹ While territorial control might be necessary in order to reach the required level of intensity envisaged by AP II and implement its provisions,Footnote ³² it is not essential for compliance with the minimum requirements of CA3.Footnote ³³

There is no reason to artificially exclude from the present study NSAGs that do not possess the level of control over the territory required by AP II. CA3 still provides a list of essential guarantees for persons detained by NSAGs in times of conflict of low intensity, while internment remains a distinctive feature of all armed conflicts regardless of their legal classification. Furthermore, as indicated above, CA3 is already limited in the scope of its application to NSAGs that are capable of implementing the basic humanitarian requirements. It necessarily means that these NSAGs, at least in principle, can ensure compliance with obligations related to the regime of detention in the situation of NIAC.

Thus, the present article seeks to explore the power to detain in respect of all NSAGs capable of constituting a party to a NIAC either under CA3 or AP II.

2.2. Definition of Detention

In the broadest sense, the notion of detention can be defined as ‘the custodial deprivation of liberty … caused by the act of confining a person in a narrowly bounded place, under the control or with the consent of a State, or, in non-international armed conflicts, a non-State actor’.Footnote ³⁴

In practice two constitutive elements of detention were elaborated:

• • the ‘objective element of a person's confinement in a particular restricted space for a not negligible length of time’;Footnote ³⁵ and

• • the subjective element, implying the lack of valid consent to the confinement in question.Footnote ³⁶

While detention is an umbrella term covering deprivation of liberty carried out by different bodies and for a broad variety of reasons, administrative detention is considered as one of its types. It is generally understood as deprivation of liberty that has been ordered by executive authorities and conducted outside criminal proceedings without the intention to bring criminal charges against the individual concerned.Footnote ³⁷

The term ‘internment’ is reserved for administrative detention carried out in the context of armed conflict.Footnote ³⁸ Despite the absence of detailed provisions specifying legal grounds and/or procedures pertaining to internment in NIAC, the term itself is mentioned in AP II.Footnote ³⁹ As such it represents ‘a specific type of non-criminal, non-punitive detention imposed for security reasons in an armed conflict’.Footnote ⁴⁰

The first distinctive feature of internment is the state of an armed conflict, which is presumed by the very use of this term introduced by IHL. Thus, some definitions used for internment in essence merely repeat the key characteristics of administrative detention.Footnote ⁴¹ The second definitive feature of internment that is frequently highlighted is the use of security grounds as a rationale for detention.Footnote ⁴² Consequently, internment could be understood as a specific type of administrative detention carried out for security reasons in situations of armed conflict.

In scholarly literature devoted to this subject such terms as ‘internment’, ‘security detention’ and ‘administrative detention’ are frequently used interchangeablyFootnote ⁴³ and will be used interchangeably for the purposes of this article. Detention carried out in the course of criminal proceedings will be referred to as ‘criminal detention’.

2.3. The Relevance of State Practice

Before exploring the issue at hand, a few methodological observations will be made regarding the overall relevance of state practice for the determination of the existence of a power to intern conferred on NSAGs by virtue of IHL.

First, in analysing the scope of powers provided by IHL, one has to keep in mind the equality of the belligerents as one of the foundations of this area of law. This principle, which implies equality of obligations, is applicable in both IAC and NIAC and is reflected in conventional provisions.Footnote ⁴⁴ Equality of belligerents is essentially grounded on the irrelevance of the underlying motivations of the warring parties in determining the scope of their obligations and corresponding isolation from the issues of jus ad bellum (for IAC) and criminal prosecution for taking up arms (for NIAC). The validity of this principle for both types of NIACFootnote ⁴⁵ and its importance for ensuring effective compliance with IHL has been confirmed repeatedly in the doctrine.Footnote ⁴⁶

In accordance with this principle, states parties to NIACs and NSAGs are accorded the same level of rights and obligations within the IHL framework. Therefore, if it is established that states are authorised to exercise administrative detention by virtue of IHL, the same authorisation is to be provided to NSAGs.

This principle, however, is not to be extended to other areas of international law. As has been reiterated on numerous occasions, equality of belligerents does not entail the overall equality of parties to a conflict under international law.Footnote ⁴⁷ This conclusion implies that NSAGs are not conferred with powers that are either based on other areas of international law or considered to be derived from the legal personality of states under the latter.

Thus, even if one accepts a hypothesis that IHL confers a limited or functional legal personality on NSAGs,Footnote ⁴⁸ it does not automatically lead to the capacity of non-state actors to participate in the formation of customary rules of international law. On the contrary, despite arguable positive implications of the acceptance of practice of NSAGs as evidence leading to the formation of a customary rule,Footnote ⁴⁹ this idea was explicitly rejected by the International Law Commission in its study on the identification of customary international law.Footnote ⁵⁰ Hence, the practice of NSAGs, although an important source of information on the implementation of IHL, cannot serve as conclusive evidence of the existence of their power to detain in NIAC. Even if consistent and conclusive practice is established on the side of the NSAG, it is unlikely to be accepted by states as law in accordance with the Statute of the International Court of Justice.Footnote ⁵¹

At the same time, we are dealing with situations where neither state is willing to take any steps or make any pronouncement that could provide even a slim chance for the legitimisation of the NSAG it is facing in the course of NIAC.Footnote ⁵² It leads to frequent denials of the existence of a NIAC from the side of the state authorities.Footnote ⁵³ As has been noted, one of the key reasons for such denial is the desire to evade equality of belligerents by staying in the realm of the law that does not put state armed forces on an equal footing with NSAGs, even if the latter implies the necessity to comply with a more strict set of obligations provided by other areas of international law.Footnote ⁵⁴

It is therefore extremely unlikely that the practice of states would provide a specific reference to the scope of rights and obligations of NSAGs under IHL sufficient to establish either the existence or the absence of the power to intern in situations of NIAC. For the authorities, the practice of detention carried out by non-state actors first and foremost constitutes an unlawful deprivation of liberty under domestic criminal law. While understandable and legitimate, it does not help in clarifying the extent of powers provided by IHL without prejudice to the qualification of the same practices within national legal systems.

Importantly, as entities NSAGs cannot possibly constitute parties to disputes before regional human rights courts or appear before either of the UN bodies entitled to consider individual complaints. At the same time, the case law of international criminal tribunals addressing administrative detention during NIACs instead of considering the issue of the existence of a right or an inherent power to intern (regardless of the nature of the detaining authorities) is concerned essentially with establishing whether the treatment of detainees was in compliance with IHL and, if not, whether such treatment amounted to a war crime.Footnote ⁵⁵ Thus, not surprisingly, the case law of international courts does not contain an examination of detention powers of NSAGs.

Taking into account the insignificance of the practice of NSAGs and the absence of relevant guidance provided by international judicial institutions, in order to establish the scope of powers granted to NSAGs one inevitably has to refer to state practice both as a source of interpretation of CA3 and AP IIFootnote ⁵⁶ as well as a constitutive element of a possible customary rule. It is true that in the vast majority of cases examples of state practice would be concerned about understanding and interpreting detention powers vested in governmental authorities in situations of NIAC. Nevertheless, such practice can provide an indication in respect of the powers the latter believe to be provided to their agents by virtue of either conventional or customary provisions of IHL.

If it is established that the power to intern is included in the above-mentioned scope of rights derived from IHL, it would necessarily mean that NSAGs also possess such power strictly within the limits provided by this specific area of law as a result of the overarching principle of equality of belligerents.Footnote ⁵⁷

Therefore, for the purposes of the present article, state practice will be analysed with the aim of establishing whether IHL provides NSAGs with a power to intern.

3.1. The Absence of Express Authorisation

One of the first problems to draw the attention of lawyers in trying to find a legal basis for detention carried out by NSAGs is the absence of express authorisation either in CA3 or in AP II. This fact was repeatedly interpreted by scholars arguing against its existence as conclusive evidence of the absence of a power to detain provided by IHL applicable in NIAC.Footnote ⁵⁸

Before turning to an analysis of the validity of this conclusion, it is worth remembering that the absence of express authorisation was never interpreted as leading to the illegality of internment under IHL as such, even by those authors who argue against the existence of any authorisation in both conventional and customary provisions.Footnote ⁵⁹ The legality of administrative detention under IHL affects the practical importance of the overall debate over the existence of the power to intern, which will be discussed later in this article.

To begin with, IHL provisions applicable to NIACs do not contain any rules that could be interpreted as a general prohibition of administrative internment. Failures in complying with obligations related to the humane treatment of detainees were repeatedly considered by the ICTY as grave violations of the principles of IHL punishable under its Statute.Footnote ⁶⁰ These violations under certain circumstances can also be considered as war crimes under the Rome Statute.Footnote ⁶¹ Despite the existence of extensive case law considering detention practices carried out during NIACs, one would not be able to find instances when the very practice of internment was interpreted as a violation of IHL.

At the same time, there are examples of practices considered legal under IHL although not expressly authorised by its provisions. For instance, the killing of members of regular armed forces by NSAGs was never considered illegal under IHL,Footnote ⁶² although neither CA3 nor AP II expressly authorise the use of lethal force by their members, even against persons that would constitute legitimate military targets during IAC.Footnote ⁶³

Furthermore, for governmental authorities an assumption of illegality of detention under IHL carried out during NIAC would mean the imposition of stricter obligations than those found in IHRL and applicable during peacetime, as the latter in their turn leave at least some room for administrative detention based on security grounds, provided that essential guarantees are respected.Footnote ⁶⁴ This conclusion would be simply illogical because, as a general rule, IHL does not further limit powers that states possess during peacetime.Footnote ⁶⁵ On the contrary, while taking into account the notion of military necessity, the IHL provisions tend to loosen the analogous restrictions imposed by IHRL.Footnote ⁶⁶

The prohibition on using riot-control agents as a method of warfare,Footnote ⁶⁷ while being an important exception to this rule, was introduced into the treaty because of a very specific rationale. Reportedly such use was perceived as creating the danger of further escalation, leading to the use of chemical weapons.Footnote ⁶⁸ Neither the text of CA3 nor AP II is indicative of a specific intent to limit the power of states to detain compared with what is already allowed under national law and relevant provisions of IHRL.

It is also impossible to discern this intention from state practice, which on the contrary tends to expand powers of state agents regarding use of lethal force and restriction of liberty through declaring a state of emergency with the eruption of a NIAC. Thus, it could be presumed that IHL at the very least would not prohibit internment as a permissible practice for state authorities during peacetime, provided that specific conditions are met. Following the above-mentioned principle of equality of belligerents, the legality of administrative detention carried out in the course of a NIAC by regular armed forces under IHL would necessarily imply the legality of the internment by NSAGs under IHL.

While assessing whether IHL contains a general prohibition of internment during NIAC, it is also important to take into account practical considerations coming from a shared interest in ensuring the compliance of NSAGs with at least minimum humanitarian requirements. The work of humanitarian organisations focusing on engagement in a dialogue with NSAGs demonstrates that these actors are willing to take ownership and responsibility for compliance with IHL provisions.Footnote ⁶⁹ There are numerous factors that can potentially make NSAGs more or less interested in abiding by relevant regulations. While the sense of responsibility, positive self-image and legitimacy in the eyes of the local population, as well as at the international level, can contribute to compliance with IHL, the lack of positive incentives coupled with criminalisation of all NSAG activity could be expected to result in decreased motivation to respect rules and further radicalisation.Footnote ⁷⁰

Thus, the automatic illegality of detention is likely to discourage NSAG members from the humane treatment of detainees. If the very fact of detention, despite being carried out in a non-arbitrary and humane way, constitutes a violation of IHL (or even worse, a war crime of taking hostages),Footnote ⁷¹ there is no sense in employing efforts in an attempt to comply with even minimum standards related to conditions of detention or legal review procedures.

Therefore, from the legal as well as from the practical standpoint the absence of an express authorisation should be interpreted as being at least neutral to the initial legal basis of internment or as implicitly authorising it.Footnote ⁷²

3.2. The Absence of Details on the Legal Procedure Established for Detention

It was also argued that the power to detain cannot be deduced from IHL provisions as the latter neither specify the scope of such power nor provide guidance on the relevant procedures.Footnote ⁷³ This conclusion, however, raises serious questions. The absence of more detailed provisions regulating the conduct of parties to a NIAC could at least be explained partly by a dilemma faced by the states that find themselves stuck between two conflicting objectives. The first of these is similar to the overall purpose of IHL and lies in the general humanitarian aspiration to provide better protection for victims of armed conflicts (including NIACs),Footnote ⁷⁴ which in its turn requires regulating the conduct of NSAGs and ensuring proper implementation of relevant provisions. It also inevitably leads to the necessity of assigning a certain level of rights and obligations. However, the second objective of states is to avoid the legitimisation of armed groups, which often results in avoiding the use of more concrete wording that could potentially be interpreted as authorising NSAGs to perform functions historically reserved solely to states.Footnote ⁷⁵ Therefore, the result we witness in AP II should be considered an inevitable compromise rather than as evidence for either side of the debate.

Given the limited conventional regulation of NIAC, its interpretation through analogous norms of GCs, customary rules or doctrine is not unusual and does not necessarily imply the absence of authorisation. In a similar way conventional regulation of NIAC does not provide us with any definition of ‘military objectives’ or any standards for determining the character of a particular object. Nevertheless, it did not lead to arguments over the power to carry out attacks against military objectives in the course of NIAC.

It was also rightly pointed out that although conventional IHL does not contain a clear definition of direct participation in hostilities (DPH) or provide any procedure for its determination either for IACs or for NIACs (not even mentioning the three-step analysis provided by the ICRC interpretive guidance),Footnote ⁷⁶ state armed forces are still considered to be empowered to attack those engaged in DPH regardless of the character of the conflict.Footnote ⁷⁷ Thus, the fact that conventional IHL does not provide specific guidance on procedures regulating the practice of administrative detention during NIAC should not be interpreted as proving the absence of power to intern. Rather than supporting any side of this debate, the absence of defined legal procedures should be treated as neutral and without prejudice to the existence of the authorisation in question.

3.3. The Possibility of Regulating Conduct Otherwise Illegal under International Law

Much of the discussion among legal scholars was focused around the question of whether IHL can regulate conduct that is illegal as such.Footnote ⁷⁸

The example of torture, which is both prohibited at all times and not regulated by IHL provisions, was used to prove that the latter cannot regulate something that is totally prohibited under international law.Footnote ⁷⁹ On the other side of the debate, the example of an initial recourse to force was used to prove that otherwise illegal conduct can be regulated by IHL without being authorised.Footnote ⁸⁰ The answer, however, does not seem to support either side of the discussion.

First, one has to consider the initial source of the illegality of a particular practice. It can be IHL itself, another area of international law or domestic law.

Turning to the first scenario, IHL cannot regulate conduct that is either prohibited by its own provisions or contravenes the very object and purpose of this area of law. In the latter situation, even in the absence of clear conventional provisions, the legality of such conduct under IHL will be effectively precluded by the Martens Clause, which establishes that in any case the population remains under the protection of ‘laws of humanity, and the dictates of public conscience’.Footnote ⁸¹ The example of torture falls within this category. This practice contradicts the essence of IHL and is also prohibited explicitly by its norms, thus it cannot be regulated. The same is true for the use of blinding laser weapons or booby traps. Clearly, internment as such is not prohibited by IHL, as it is applicable in NIAC, and does not fall within this scenario.

We will now turn to illegality based on either domestic law or provisions of international law (other than IHL). IHL has been developed intentionally in a way that does not depend on legality or illegality of the use of force under jus ad bellum.Footnote ⁸² The same is true for its interrelationship with national legal systems. Special caution was exercised during the negotiation process to clear AP II from any terms that could even slightly suggest any special status or recognition being granted to NSAGs by virtue of the treaty.Footnote ⁸³ It also does not preclude state parties from prosecuting insurgents for taking up arms and using lethal force against members of regular armed forces under the national criminal law, even though their conduct might be perfectly legal under IHL.Footnote ⁸⁴

Therefore, if the question is whether IHL can potentially regulate acts that under normal circumstances constitute a violation of domestic law or contravene other areas of international law (for instance, jus ad bellum or IHRL) – it certainly can. Following the same logic, IHL can regulate detention by NSAGs (otherwise illegal under the national law) without authorising it.

In the end, applying these two scenarios to our situation, one has to conclude that the mere fact of regulation of detention cannot be used as conclusive evidence of its authorisation by virtue of IHL. Nor it can be used as evidence of its absence. Unfortunately, it does not clarify the matter in question in any way.

3.4. The Prohibition of Arbitrary Detention

One of the reasons why this issue has attracted the attention of so many scholars in the first place is the general prohibition of arbitrary detention under the provisions of IHRL.Footnote ⁸⁵ The unanimous position of all relevant human rights bodies (regardless of the specific instruments that led to their establishment) supports the validity of this prohibition during armed conflict.Footnote ⁸⁶ The customary status of this prohibition has also been confirmed by various bodies.Footnote ⁸⁷ The problem lies in one of the elements of the test set by the human rights instruments in defining what constitutes arbitrary detention and consequently what does not. It is the requirement of detention being imposed and effected ‘in accordance with a procedure prescribed by law^’.Footnote ⁸⁸ Reconciling this requirement of a valid legal basis with the absence of an explicit authorisation being provided by conventional provisions of IHL has proved to be a difficult task.

According to some commentators who argue against authorisation to detain in NIAC being provided by virtue of either conventional or customary IHL, detaining authorities have to establish a legal basis for internment elsewhere; otherwise it will be considered arbitrary and thus in violation of international law (meaning IHRL), regardless of its grounds and compliance with corresponding procedures.Footnote ⁸⁹ However, declaring internment that is not authorised by virtue of either domestic law or some other explicit provisions of international law as arbitrary is problematic on several grounds.

As clarified in the preceding section, detention in general can be regulated by IHL even if it is considered to be a violation of rules belonging to other areas of international law. However, the problem is that arbitrary detention is prohibited in all types of conflict by IHL itself and, especially in prolonged cases, it can also amount to inhumane treatment, which in turn constitutes a serious breach of IHL's own provisions.Footnote ⁹⁰

Then, if we consider arbitrary deprivation of liberty to have exactly the same legal meaning in IHL as it does in IHRL provisions, the absence of authorisation to detain by the former will render all internment carried out by NSAGs as arbitrary and thus illegal. This hypothesis will also lead to the illegality of the majority of internment practices carried out by states in extraterritorial NIACs. In the latter situation, unless a clear legal basis is provided by a United Nations Security Council (UNSC) resolution or national legislation of the state in which a NIAC takes place, the legal basis requirement most probably will not be met either.

Therefore, if we conclude that there is no legal basis for internment in NIAC provided by IHL, while arbitrary detention has the same meaning in the two areas of law, we will inevitably face a dilemma that either arbitrary deprivation of liberty is not outlawed in the situation of NIAC or that IHL applicable to NIAC indeed regulates detention, which for at least 50 per cent of detaining authorities is illegal per se under their own provisions.Footnote ⁹¹

It is worth noting that the first option was not supported even by those scholars who argue for the absence of either explicit or implicit authorisation to detain in IHL, and thus the general illegality of internment during NIAC, unless the legal basis requirement could be satisfied by other areas of law.Footnote ⁹² The second proposition brings us back to the analysis provided in Sections 2.3 and 3.3. IHL by its nature was intended to regulate the conduct of NSAGs as much as the conduct of states. The illegality of all detention carried out by NSAGs would then necessarily mean that IHL provisions, and CA3 in particular, were drafted to regulate the conduct they outlaw. Such an interpretation would not only be illogical; it would also contradict the predominant practice of non-regulating conduct that is outlawed by means of the same instruments of IHL.

There are two possible ways of avoiding this dilemma. The first implies that IHL provides a legal basis for internment and prohibits arbitrary detention, applying the same legal standard as relevant human rights bodies.Footnote ⁹³ The second solution is to accept that arbitrary deprivation of liberty has its separate meaning within IHL, which does not specifically require detaining authorities to establish a concrete legal basis.

In the author's opinion, it is far more logical to accept the second solution. When trying to unpack the notion of arbitrary detention within the framework of IHL, attention needs to be given to the initial basis of its prohibition as well as the overall history of the development of this area of law.

The prohibition of arbitrary detention by virtue of CA3 can be dated back to 1949, before the adoption of the majority of human rights instruments and certainly long before the jurisprudence of various human rights bodies affirmatively establishing legal basis as one of the relevant criteria. Thus, it would not be surprising if this prohibition within the IHL framework, despite pursuing the same goal of protecting detainees, followed its unique path of development leading to a different understanding of arbitrariness.

The initial idea standing behind the development of CA3 was the willingness to protect victims of NIAC by extending the application of the key principles and safeguards of the GCs to this type of conflict.Footnote ⁹⁴ The resolutions of the International Red Cross Conference evidence the fact that rather than replicate human rights standards (which by then had developed within national legal systems), states were gradually coming to extend the application of IHL principles developed in earlier international instruments belonging to the same area of law.Footnote ⁹⁵ Therefore, in order to define the scope of the prohibition of arbitrary detention during NIAC one would have to analyse relevant provisions of the GCs and customary IHL that preceded their adoption.

The idea of restricting the permissible grounds of detention for the purposes of IHL dates back to the nineteenth century. Looking into historical instruments and well-known military manuals that codified existing customs regulating the conduct of hostilities, we can identify two types of detention considered to be in compliance with IHL: (i) status-based detention of PoWs,Footnote ⁹⁶ and (ii) detention of civilians justified by security needs of the warring parties.Footnote ⁹⁷ Interestingly, as early as 1880, the Oxford Manual provided that ‘the reasons justifying detention of the captured enemy exist only during the continuance of the war’.Footnote ⁹⁸

These ideas were formulated in the GCs, making non-arbitrariness of detention essentially dependent on being ‘necessary as a result of war’Footnote ⁹⁹ and lasting for as long as the initial reasons for the internment existed.Footnote ¹⁰⁰ The only difference comes from the fact that such necessity is presumed for the duration of hostilities in respect of members of the armed forces of an enemy,Footnote ¹⁰¹ while internment of members of the civilian population needs to be justified by specific risks for the security of the detaining power.Footnote ¹⁰² ‘Being necessary as a result of war’ was also justified as a qualifying criterion for non-arbitrariness in literature based on textual and historic interpretation of CA3 and Article 27 of GC IV.Footnote ¹⁰³

The recently developed Copenhagen Principles and Guidelines, following the same line of reasoning, formulated non-arbitrary detention as being carried out based on ‘valid reasons that are reasonable and necessary in light of all the circumstances’ and serving ‘a lawful and continuingly legitimate objective’.Footnote ¹⁰⁴

Combining all elements, it could be concluded that IHL does indeed provide its own understanding of arbitrariness, which requires valid administrative detention to:

• • be necessary as a result of armed conflict;

• • be justified by an objective threat to the security of the detaining power;

• • last strictly for as long as the reasons that necessitated the internment exist.

The fact that arbitrary detention has a separate meaning within the framework of IHL could also find its support in the practice of regional human rights bodies.Footnote ¹⁰⁵

The existence of two separate meanings of arbitrariness does not represent an answer to all questions as the international law, despite being increasingly prone to fragmentation, does not provide for a clear and uniform solution for a conflict between legal rules belonging to two different specialised and autonomous areas of law.Footnote ¹⁰⁶ Therefore, the non-arbitrariness of internment conducted during NIAC and its legality under IHL (provided it complies with relevant standards and procedural requirements) does not necessarily lead to the legality of the latter under international law in general, as IHRL would still require a legal basis for administrative detention for the latter to be in compliance with its provisions. In the end, the prohibition against arbitrary detention, as much as the previously considered arguments, does not evidence either the existence or the absence of the power to detain provided by IHL.

3.5. (Non-) Confirmative Practice of States

If we set aside for a while the scholarly debates regarding the nature of authorisation and direct our attention to the realities of contemporary NIAC, we find that the overwhelming practice of both states and NSAGs is to use internment as a valid security measure and a preferable alternative to killing fighters belonging to the opposite side as well as civilians who represent a security threat for a warring party. The use of internment by all states parties to armed conflicts was one of the reasons that led the ICRC,Footnote ¹⁰⁷ as well as some independent scholars,Footnote ¹⁰⁸ to conclude that an inherent power or authorisation can be deduced from customary IHL. The same conclusion was supported in the updated commentary on GC I,Footnote ¹⁰⁹ which aimed in particular to review provisions of IHL against the practice of their implementation since 1949.

Despite the existence of such overwhelming practice, its value was heavily disputed by some commentators, claiming that it was not substantiated by opinio juris and was, thus, irrelevant for the formation of customary IHL that authorises internment.Footnote ¹¹⁰

One argument that is frequently voiced is the following. As in the majority of cases states tend to rely on other areas of law in claiming their power to detain, it can be presumed that IHL is not considered to authorise detention in NIAC.Footnote ¹¹¹ This conclusion is far from being unproblematic. First, it is worth noting that in the situation where several branches of law can potentially provide the authorisation to detain, states are free to rely on the one they deem the most convenient or appropriate in the given circumstances. It does not necessarily mean that states do not believe that IHL does contain such authorisation.

The legal grounds frequently cited by states when referring to internment in NIAC are (i) domestic lawFootnote ¹¹² and/or (ii) UNSC resolutions.Footnote ¹¹³

Reliance on domestic law in such circumstances seems the most reasonable given the nature of NIAC, the monopoly of state institutions to use force within state borders and the existence of detailed regulation in this respect. For most of the time states simply do not need to search for their power elsewhere, unless we are talking about extraterritorial NIAC or situations where states have to justify their acts before human rights bodies because of their failure to derogate from the relevant IHRL provisions.

The second reason relates more to political considerations of states engaged in NIAC on their territory. The overwhelming majority of governmental authorities are reluctant to admit the actual existence of hostilities that amount to a NIAC under CA3 let alone AP II, which sets the territorial control by the NSAG as a prerequisite for such qualification.Footnote ¹¹⁴ Not surprisingly, the failure of the state to admit to an armed conflict leads to inability and/or unwillingness to rely on IHL provisions for the purposes of establishing a legal basis for detention.

The reliance of states on UNSC resolutions, rather than IHL, can also be explained in several ways. First, the supremacy of state obligations under the UN Charter provides a tempting opportunity to avoid the potential conflict of norms between IHL and IHRL.Footnote ¹¹⁵ While this conflict is by no means new, in the absence of clear guidance on the rules that should ultimately be applied in each concrete situation, complemented by the complexity of the lex specialis doctrine and inconclusive case law varying from one human rights body to another, references to UNSC resolutions can be considered a far safer basis for claiming authorisation.Footnote ¹¹⁶

The next potential reason can be found in the overall principle of equality of belligerents embodied in IHL, thus granting equal powers to NSAGs. As indicated in Section 2.3, states tend to avoid such discussions and arguments as much as possible and therefore might search for authorisation in the areas of law that provide such powers solely to states.

The third reason relates particularly to those countries falling within the jurisdiction of the European Court of Human Rights (ECtHR) as its jurisprudence is cited frequently in literature in support of the lack of opinio juris.Footnote ¹¹⁷ In several cases considered by the ECtHR, the United Kingdom preferred to rely on UNSC resolutions as a legal basis for administrative detention carried out in violation of the ECHR rather than referring the Court to rules of IHL.Footnote ¹¹⁸

In this situation reference to UNSC resolutions could potentially be explained by the fact that until the Hassan caseFootnote ¹¹⁹ the ECtHR had never referred directly to IHL provisions. Even when the Court considered acts of states that under normal circumstances would be assessed solely in the framework of IHL, as lex specialis sidelining the human rights provisions – for instance, the proportionality of aerial bombardment in the context of a de facto NIAC – it still avoided any references to IHL.Footnote ¹²⁰ Unlike the American Convention on Human Rights, the ECHR does not provide the same room for interpretation of what constitutes ‘arbitrary detention’ in the context of armed conflict, as internment does not fit within the narrow list of permitted grounds for detention formulated in Article 5.Footnote ¹²¹ Such a formulation, combined with persistent reluctance to treat IHL as lex specialis (until the Hassan case)Footnote ¹²² did not provide incentives to rely on the latter as a legal basis for internment.

It could fairly be noted that potential reasons for the reluctance of states to rely directly on IHL provisions constitute merely plausible assumptions. It is not submitted that these reasons necessarily explain the conduct of all states searching for a legal basis for internment. However, they are no less plausible than the argued assumptions of the absence of opinio juris justified by states referring to national law and/or UNSC resolutions.

It is also important to note the emergence of examples of state practice, which indicate IHL specifically as the legal basis authorising internment.Footnote ¹²³ Apart from individual positions, the same approach was expressed by participants of the Chatham House expert meeting organised by the ICRC, which confirmed the existence of an inherent power or ‘qualified right’ to intern.Footnote ¹²⁴ An analogous conclusion can be also drawn from the Copenhagen Process Principles and Guidelines in providing guidance on the humane treatment of detainees in the framework of extraterritorial NIAC.Footnote ¹²⁵ While it was drafted very cautiously, indicating specifically that no new obligations or authorisations were brought into force by the virtue of its provisions, this document is based on an important presumption that such detention can be conducted in a non-arbitrary way and in accordance with international law.Footnote ¹²⁶

It is worth keeping in mind that in the majority of extraterritorial NIACs the legal basis requirement could be satisfied only if the latter is found in IHL unless detaining authorities receive direct authorisation in the national law of the territorial state. The practice of the ECtHR indicates the unlikeliness of UNSC resolutions being accepted as justification for departing from obligations set out in human rights instruments.Footnote ¹²⁷ On the other hand, it is exactly IHL provisions treated as lex specialis that led the ECtHR in its ruling in Hassan Footnote ¹²⁸ to depart from its long-standing reluctance to admit the legality of internment under the ECHR in the absence of a lawful derogation.Footnote ¹²⁹

Therefore, it is the opinion of the author that the presumption of the possibility of administrative detention being carried out during extraterritorial NIAC in a non-arbitrary way and in compliance with international law would necessarily lead states to admit that IHL (either conventional or customary) does authorise internment and provides the legal basis essential for compliance with relevant human rights instruments. In turn, it would inevitably lead to an implicit admission of authorisation of internment to be extended to NSAGs based on the equality of powers of the parties to NIAC under IHL.

3.6. Legality or Authorisation?

The last argument to be considered in this section can be expressed in the following way. The overwhelming practice of internment on behalf of all parties to NIAC is merely evidence of the legality of internment under IHL as opposed to implicit authorisation. Here we are turning to a very interesting issue. It seems that in many cases the position on the existence of authorisation to intern within the framework of IHL (in particular, when considering NSAGs) becomes a matter of personal belief and the legal debate turns into an exercise that does not affect practice in any meaningful way.

While states are not sufficiently vocal (for various reasons indicated in earlier sections) in demonstrating opinio juris beyond any doubt, and the drafting history of CA3 and AP II does not provide sufficient guidance regarding the initial intention of states, the position of scholars to a large extent depends on what they themselves consider to be the implied content of the relevant articles and, ultimately, what would be in the best interest of detainees.Footnote ¹³⁰

In order to assess the overall relevance and practical implications of the debate on the existence of authorisation to detain we have to return to one of the starting points of the present analysis elaborated in Section 3.1: the legality of administrative detention under IHL, provided that the detention complied with certain procedural requirements.Footnote ¹³¹

For the purposes of IHL, it does not matter what the source of this legality is: direct authorisation, the absence of prohibition, or acquiescence in existing practice.

There are three possible scenarios:

• • IHL authorises and provides a legal basis for detention in NIAC. IHL contains a similar legal test for arbitrary detention encompassing the legality principle.

• • IHL does not authorise detention in NIAC, although detention is still legal under its provisions. IHL contains its own legal test for arbitrary detention, which does not require explicit authorisation provided by law.

• • IHL does not authorise detention and it is therefore arbitrary and illegal under its own provisions.

The last scenario is not supported by either side of the debate. Furthermore, it is illogical and entails dangerous practical consequences for general IHL compliance related in part to the humane treatment of detainees. As for the first two scenarios, there is hardly any difference for NSAGs for several reasons.

In both cases, administrative detention as such would not constitute a war crime.Footnote ¹³² Therefore, it would not lead to automatic criminal prosecution and would leave the door open for amnesties for members of NSAGs at the end of the NIAC.Footnote ¹³³ Furthermore, the absence of the label of ‘war criminal’ for NSAGs that duly complied with IHL provisions regarding the treatment of detainees can potentially assist them in gaining international legitimacy.

Both cases also reflect the principle of equality of belligerents,Footnote ¹³⁴ as in the majority of cases state forces will be legally detaining members of NSAGs authorised by national legislation. All of these factors, coupled with a slim possibility of obtaining an amnesty in the aftermath of an armed conflict, create an important incentive for enhancing compliance with IHL on behalf of NSAGs. The first two scenarios thus potentially entail the same practical consequences.

They are also equal for the purposes of criminal prosecution. Crimes committed by members of NSAGs are prosecuted either within the domestic judicial system or before international criminal institutions.

In normal circumstances internment by NSAGs will inevitably constitute an illegal deprivation of liberty that infringes state monopoly to use force and entails criminal responsibility under a national penal code, once again regardless of the existence of authorisation in question. It has always been a distinctive characteristic of the regulation of NIAC by IHL: its provisions do not in any way preclude the nation-state from prosecuting members of NSAGs for acts that are considered lawful or authorised or tolerated by IHL. Such acts include, in particular, attacks on members of armed forces and deprivation of liberty carried out by non-state actors. Both could be carried out in compliance with IHL while still constituting a violation of domestic law. Therefore, neither the absence nor the presence of a legal basis for internment carried out during NIAC is likely to dramatically affect the outcome of judicial proceedings against members of NSAGs at the national level.

International criminal institutions will consider individual violations in the framework of international criminal law based on their jurisdiction as defined in their statutes (for ad hoc bodies) or the Rome Statute for the International Criminal Court. The jurisdiction of such institutions, especially in relation to NSAGs, is first and foremost focused on violations of IHL in the framework of armed conflict. Considering the overall legality of internment under IHL applicable to NIAC, the very fact of internment will not constitute a war crime provided that the detaining powers comply with the relevant procedural requirements and responsibilities related to humane treatment.Footnote ¹³⁵

At the same time, when addressing crimes against humanity, these courts or tribunals tend to set the bar for criminal responsibility extremely high in order to deal with the most atrocious violations, for which no authorisation could be found in IHL.Footnote ¹³⁶ It is extremely unlikely that internment carried out by NSAGs in compliance with IHL would amount to a widespread or systematic attack directed against the civilian population in violation of fundamental rules of international law only by virtue of non-compliance with the legal basis requirement set by IHRL.Footnote ¹³⁷

Therefore, it can be concluded that regardless of how interesting and intellectually challenging the debate on the power to detain might be, it seems to have very limited practical importance for assessing the conduct of NSAGs both during the conflict and in post-conflict settings.

The only area of law for which the existence of an authorisation to detain might indeed be relevant and entail concrete practical consequences is IHRL. However, it seems that regardless of whether we agree or disagree with the fact that IHL contains implicit authorisation to detain, or whether its legal test for non-arbitrariness of detention does not include the principle of legality, it is still up to human rights bodies to consider taking into account all the circumstances of a particular case, if provisions of IHL constitute lex specialis.Footnote ¹³⁸

Even if IHL does contain authorisation or a ‘conditional right’ to detain, it is questionable whether it will result in an automatic settlement of the conflict of norms, especially in the absence of lawful derogation from relevant provisions of international and/or regional human rights instruments. After all, despite the presence of an explicit authorisation for administrative detention in IAC provided by GC IV, the rigid formulation of Article 5 of the ECHR, with a defined list of exceptions, led to a repeated denial by the ECtHR to accept internment as a lawful practice.Footnote ¹³⁹ That is why the judgement in Hassan Footnote ¹⁴⁰ – which not only referred directly to IHL as a permissible source of interpretation of obligations under the ECHR but also allowed such an interpretation in the absence of a lawful derogation – represented a significant deviation from the earlier jurisprudence of the ECtHR. This has caused much discussion among legal scholars.Footnote ¹⁴¹

The practice of the application of IHL within the Inter-American system could also be described as non-uniform, varying from one body to another. Thus, while the Inter-American Commission of Human Rights found itself able to apply provisions of IHL, when appropriate, to a case in question,Footnote ¹⁴² this position was not shared by the Inter-American Court of Human Rights.Footnote ¹⁴³

Aside from the disputed legal basis of internment in situations of NIAC one could notice more general hesitation of human rights bodies to apply IHL provisions. This could be for various reasons, starting from the fact that direct application of IHL was not envisaged by states when providing these bodies with jurisdiction, to reluctance to reduce the level of protection granted by IHRL, especially when state authorities not only repeatedly fail to derogate from provisions of the latter, but frequently deny the very existence of a NIAC on their territory.

Ultimately, the existence or absence of authorisation to detain based on IHL is particularly relevant only for the states parties to human rights instruments, because even if one concludes that NSAGs have limited legal personality and certain obligations under IHRL, they are still incapable of becoming a party to a case that is dealt with by any human rights institution.

It is the opinion of the author that the only serious implication of the choice we make between authorisation and legality lies in regulation of the conduct of state armed forces during extraterritorial NIAC. In reality, it is the only scenario where the absence of authorisation and the mere legality of internment under IHL might affect the jurisprudence of human rights institutions, which will be more inclined to find a state party to a particular instrument in violation of its obligations (especially in the absence of a lawful derogation), given the scarcity of conventional safeguards protecting the rights of detainees.

It is states that are primarily interested in interpreting IHL provisions as authorising detention. Therefore, the growing involvement of state armed forces in extraterritorial NIAC and the scrutiny of regional human rights institutions in monitoring detention practices is most likely to lead to governments being more vocal about the existence of such authorisation. Thus, we might find more evidence of opinio juris in the coming decades.

It is the opinion of the author that none of the arguments expressed by either side of the debate provide evidence beyond a reasonable doubt of the absence or of the existence of authorisation to detain. The practice of states that have only started to face challenges of an ambiguous legal basis of internment in extraterritorial NIAC is also inconclusive, despite certain indications that point to the existence, or at least the gradual formation, of a customary provision.

In the meantime, one can conclude that internment as such, if carried out during NIAC regardless of the character of the detaining authority, is legal under IHL provided that it complies with the relevant procedural requirements. Such internment is based on the existence of an inherent power to detain. It does not mean, however, that the existence of an implicit inherent power would satisfy the legal basis requirement under IHRL or otherwise change the qualification of internment under other areas of international law and/or national law. Rather, this legality and inherent power exist within the framework of IHL and for the purposes of the latter.

The debate we are experiencing today in respect of the power to detain is similar to that relating to the source of the obligatory nature of CA3 and AP II for NSAGs. Even today, distinguished legal scholars disagree on whether it is their customary nature, national implementation, direct application or some other reason that makes IHL provisions obligatory for NSAGs. However, what is important is that they would, with a rare exception, agree on their binding character.

The evolvement of state practice and international jurisprudence on the existence of the power to detain is most likely to lead to the same result. The international legal community might sooner or later be forced to accept that administrative detention in NIAC is indeed authorised by IHL without necessarily agreeing on the source of such power.

Given the limited practical relevance of the ultimate choice between the legality and authorisation of internment carried out by NSAGs, it is justified to leave the matter of the present debate unresolved, while focusing the efforts of scholars and practitioners on determining the legal framework and content of procedural safeguards applicable in the situation of internment during NIAC. The results of this theoretical endeavour are more likely to play a decisive role in the enhancement of protection of detainees and make a difference in the implementation of IHL.

4.1. Inadmissibility of Interchangeable Use of Distinct Grounds for Detention

Turning to detention itself when carried out by NSAGs during NIACs it is important to distinguish between two different types of detention: (i) administrative detention, intended to eliminate security threats posed by the detainee and, to a large extent, mirroring the internment of prisoners of war and members of the civilian population under GC III and GC IV, and (ii) criminal detention in the framework of the prosecution of violations of IHL.

It is important that these types of detention are not confused or used interchangeably. It is true that members of the opposite side could be captured and detained in order to avoid their return to combat, but then subsequently prosecuted for violations of IHL if there is evidence that such violations had taken place. However, internment should not be a substitute for the actual prosecution.

It has been proposed in the literature that internment can be used equally as an alternative to criminal prosecution when NSAGs are not capable of complying with fair trial requirements as provided by Article 6 of AP II, Article 75 of AP IFootnote ¹⁴⁴ (as a norm of a customary nature) and relevant human rights provisions.Footnote ¹⁴⁵ This position, though seemingly tempting at first sight, raises serious concerns.

First, the internment has to be based on specific grounds (which will be discussed later), which are (a) distinct, and (b) not necessarily present in a specific case of an individual facing criminal prosecution. It is true that if we are talking about active servicemen who might potentially rejoin their forces, the existence of a security threat for the NSAG tends to be presumed. If we consider a similar distinction between internment of civilians and PoWs during IAC, while the first is an exceptional measure requiring a clear security ground to be present and reviewed before an impartial and independent body,Footnote ¹⁴⁶ GC III on the contrary starts from a presumption of the right of state parties to a conflict to intern members of the opposing armed forces until the end of active hostilities or, alternatively, until they no longer represent a threat as a result of their medical condition.Footnote ¹⁴⁷

However, even if one adopts the same presumption for members of state armed forces captured by NSAGs during NIAC, the replacement of criminal prosecution with administrative internment is undesirable. The latter measure is not punitive in its nature and must be lifted as soon as security grounds for the initial decision are no longer valid. If administrative internment replaces criminal prosecution it becomes a punishment imposed without proper guarantees of an impartial adversarial process and an opportunity to challenge the charges. Apart from being unfair, such replacement bears additional risks of inhumane treatment of a detainee (as a result of perceived validity of the charges) and indefinite detention caused by the inability to challenge its grounds and potential unwillingness to release a detainee even after the security risks have disappeared. For instance, a detainee might cease to pose a threat because of a change in his or her medical condition or injuries incurred in the course of an attack preceding capture, but the NSAG authorities might still be unwilling to release this person if the term of detention is considered to be insufficient as a punishment for a crime this person has allegedly committed.

Furthermore, the same group might be willing to prosecute a civilian who has allegedly committed a war crime in the course of direct participation in hostilities, or even a common law crime committed on the territory controlled by a particular NSAG. Leaving aside the question of whether the NSAG is permitted to exercise such power on the territory under its control, in this case a properly exercised criminal prosecution becomes the only possible legal basis for continuing detention. As in the situation described above, such individuals would be practically denied their right to an effective and impartial review. Indeed, the absence of a proper investigation, lack of access to evidence of guilt and legal review of the latter together deprive detainees of an opportunity to challenge their status in a meaningful way.

4.2. Permissible Grounds for Internment

Neither CA3 nor AP II provide guidance on what would be a permissible ground for internment. However, the rationale behind the use of such a measure during NIAC is inherently the same as it is in IAC and is driven by what ‘may be necessary as a result of the war’.Footnote ¹⁴⁸ It seems only reasonable, therefore, that the regulation incorporated in the GCs can provide useful guidance for NIACs. In particular, reference can be made to Article 42 of GC IV, which stipulates that ‘[t]he internment … may be ordered only if the security of the Detaining Power makes it absolutely necessary’.Footnote ¹⁴⁹

Article 78 of GC IV, which regulates the relationship between the occupying power and the civilian population, provides a similar standard to be followed: ‘If the Occupying Power considers it necessary, for imperative reasons of security, to take safety measures concerning protected persons, it may, at the most, subject them to assigned residence or to internment’.Footnote ¹⁵⁰ Despite certain differences in the language used, both articles establish similar grounds for detention: the existence of a threat to security based on an objective examination of facts.

A purely subjective assessment of a perceived threat would constitute arbitrary detention falling within the prohibition of inhumane treatment and the overall prohibition of arbitrary detention under IHRL as such, which does not lose its validity during armed conflictsFootnote ¹⁵¹ as well as during any other public emergency.Footnote ¹⁵² Not surprisingly, imperative reasons of security were upheld as a proper standard in the updated commentaries on the GCs in determining a test to assess the legality of administrative detention under CA3.Footnote ¹⁵³

It is also vital to keep in mind that the above-mentioned threat must be posed by a specific individual whose internment was ordered by the NSAG. Although it does not come from the literal reading of relevant articles, this conclusion can be deduced from the same prohibition of arbitrary detention. For instance, if there is intelligence confirming that the attack against the NSAG was carried out by a resident of a particular village, it will not be justified to intern all male members of this community. This measure would not be absolutely necessary as the majority (or all) of those detained by default do not pose any security threat. This detention would also acquire an inherently punitive nature, prohibited under Article 4(2)(b) of AP II.Footnote ¹⁵⁴

Again, similar to the regulation of IAC, one could expect that the so-called ‘status-based’ detention would be exercised in situations of NIAC.Footnote ¹⁵⁵ Although IHL does not provide a clear combatant versus civilian distinction for NIACs, we can still identify two categories of person whose loss of protection against an attack is permanent, unless he or she turns hors de combat. These categories include members of state armed forces (in the case of a NIAC with the participation of at least one state) and members of NSAGs with a continuous combat function.Footnote ¹⁵⁶ Although the ICRC Interpretive Guidance has been criticised repeatedly for adopting formulations that were too narrow,Footnote ¹⁵⁷ such narrowness is aimed at protecting the rest of the population who do not pose any permanent threat for the security of parties to a conflict or inflict harm. These persons, therefore, do not lose their immunity on a permanent basis and cannot be attacked at any time, unlike the categories described above. The same distinction is relevant for the purposes of detention. Status-based detention would presume the existence of a security threat posed by members of the state armed forces and/or members of NSAGs exercising a continuous combat function (although it would not deprive them of their right to contest this presumption in the process of a legal review). The rest of the population, including civilians engaged in direct participation in hostilities before their capture, could be interned only if a specific security threat is demonstrated before the reviewing authority.

The idea that grounds for internment by NSAGs should somehow be interpreted in a stricter way compared with those applicable to governmental detaining authorities – for instance, by not allowing non-state actors to intern civilians based on security grounds – raises serious concerns.Footnote ¹⁵⁸ This argument is not based on either conventional or customary provisions. First, it seems that these arguments contravene the basic principle of equality of belligerents, thus creating a disincentive for NSAGs to comply with what is perceived to be inherently unjust rules. In addition, the inability to detain based on real (and sometimes legitimate) security concerns might lead to carrying out ‘fake’ criminal prosecutions of such individuals. It goes without saying that the reverse substitution of administrative internment with detention in the course of criminal proceedings is equally dangerous, not only because of the unfairness of the whole procedure but also because it deprives a person detained from periodic review that could lead to his or her release when the security concerns are no longer valid.

4.3. Procedural Requirements

In respect of procedural requirements regulating administrative detention, the situation is perhaps even less clear than with the initial grounds of internment, given that the absence of specific legal provisions is topped by limited capacities of NSAGs to establish proper institutions and procedures. Regardless of how vigorous and intellectually challenging the debate is on the issue of initial authorisation to intern, detention in NIAC will remain a common practice carried out by both parties to a conflict. It will also remain a practice that leads to extreme vulnerability of hundreds of thousands of people, exacerbated by the absence of clear regulation. It is thus argued that the key concern of the international legal community should shift from the ‘power to detain’ to the elaboration of clear and practical standards of protection of detainees in NIACs and possible measures to enhance their implementation.

In the absence of a legal framework provided by either CA3 or AP II, procedural safeguards limiting the powers of NSAGs as detaining authorities are to be derived from the general prohibition of arbitrary detention. As was established in Section 3.4, IHL encompasses its own understanding of non-arbitrariness of detention, based on the ability of the detaining authorities to demonstrate that internment is necessitated by ongoing armed conflict, justified by a specific security threat posed by an individual, and ends at the moment the established threat ceases to exist. Therefore, procedural guarantees must provide reasonable means to verify the validity of grounds of internment and their continuing relevance for the duration of the armed conflict.

In order to establish a concrete form that these safeguards should take, one can refer to relevant provisions of IHL regulating IAC. As the overall intention of the drafters of the GCs, including CA3, was to extend the application of humanitarian principles applicable in situations of IAC to NIAC,Footnote ¹⁵⁹ it seems only reasonable that guarantees established for internees captured by state forces can be applied by analogy to persons detained by NSAGs facing the same risks and challenges (provided that these guarantees are not dependent on privileges and immunities associated with the status of PoWs).

Procedural standards established by relevant human rights bodies and constituting core protection against the arbitrariness of detention can be used as a supplementary means of interpreting guarantees established by IHL. Although these areas of law do not necessarily contain the same understanding of arbitrary detention, IHL and IHRL require a similar set of procedural safeguards to verify the character of such internment. Therefore, it is important to take into account guarantees elaborated in the practice of the application of IHRL, which continue to apply in times of armed conflict and emergencies.Footnote ¹⁶⁰

The following minimum requirements can be established:

• • An individual is to be informed of the reason for his or her detention. This requirement is established by the majority of human rights instrumentsFootnote ¹⁶¹ and is included in the list of minimal guarantees provided by Article 75 of AP I and recognised for their customary nature.Footnote ¹⁶² Regardless of whether one prefers to apply relevant IHL provisions on internment in IAC by analogy, or define this procedural guarantee via human rights standards that are equally applicable in times of NIAC, the latter constitutes an essential part of fair trial procedures and thus will be required from the detaining authority. Lack of information as to the grounds of the detention will also inevitably render such detention arbitrary and thus in violation of core customary provisions of IHL.

• • Procedures of internment must allow the grounds of detention to be challenged before an impartial, independent and competent body. The opportunity to challenge the legality of one's detention is considered one of the fundamental guarantees by the major human rights instruments.Footnote ¹⁶³ Whereas IHL does not require such review to be performed by a judicial body, it is essential that the reviewing body functions in a way that allows detainees to challenge their detention in a meaningful way. It is true that these standards (as far as NIAC is concerned) are set primarily with respect to regular judicial proceedings.Footnote ¹⁶⁴ However, no rationale can be found for internees whose detention might last for as long as that of a criminal suspect, especially in the context of protracted conflicts, to be subject to a lesser degree of protection. The applicability of this safeguard has been further confirmed by various human rights bodies.Footnote ¹⁶⁵ Its focus is not on the formal organisation of the reviewing body but rather on the way in which it functions in practice and whether its procedures allow an efficient review. The review can be conducted by an administrative body as well as by military commissions.Footnote ¹⁶⁶ It is vital, however, that members of such bodies (i) did not order the internment in the first place; (ii) are independent from those who sanctioned the internment as well as from the general command structures, and are not appointed by the latter;Footnote ¹⁶⁷ (iii) do not perform active military duties;Footnote ¹⁶⁸ and (iv) are empowered to order immediate release.Footnote ¹⁶⁹

• • The review is to be conducted periodically.Footnote ¹⁷⁰ It is obvious that circumstances change in the course of time and a detainee may no longer pose a threat to security, resulting in his or her detention becoming arbitrary. Thus, even in the absence of a clear conventional provision, the failure to provide an opportunity for such review will inevitably entail a violation of the prohibition of arbitrary detention by NSAGs.

There are several other procedural rights that can provide NSAGs with useful guidance for conducting fair proceedings, such as provision of legal assistance and adequate means of defence (including time and space), and the right to attend the hearing of the review body. These rights were mentioned by Jelena Pejic as being applicable to situations of administrative internment.Footnote ¹⁷¹ It is the opinion of the author, however, that these guarantees (while certainly desirable) are not provided by IHL at the current stage of its development. While these safeguards can be considered as important factors contributing to the overall impartiality of the process, one cannot find any conventional provisions that would make them obligatory, even for internment during IAC. At the same time, as will be elaborated in the next section, the uncertainty regarding the applicability of IHRL to actions of NSAGs precludes the imposition of human rights standards by analogy when the latter are not replicated or otherwise based on IHL.

The lack of capacity of NSAGs to uphold the same standards has even led to a debate on whether one should introduce the sliding scale of obligations applied to NSAGs.Footnote ¹⁷² When addressing procedural requirements accompanying the process of reviewing validity, we should nevertheless remember that these provisions constitute the very core, minimal standard of what is essential for the protection of civilians against arbitrary deprivation of liberty during NIAC. This standard is already significantly less demanding compared with detailed provisions of IHRL, and further ‘accommodating’ it to the specific institutional characteristics of NSAGs might lead to devaluation of its initial objectives.

Nevertheless, further clarification of applicable legal standards is crucial for improving the implementation of IHL on the side of NSAGs. While there are numerous factors that can lead to potential non-compliance by these actors with relevant IHL provisions,Footnote ¹⁷³ it is proven empirically that a significant number of NSAGs are interested in abiding by international law as long as such compliance brings international legitimacy.Footnote ¹⁷⁴ Some of these groups that have existed for decades have proved to be successful in developing institutional capacities sufficient for compliance with the basic humanitarian requirements established by IHL.Footnote ¹⁷⁵ Clarification of the legal framework can provide NSAGs that are interested in gaining international legitimacy with clear guidance, which would ultimately serve the interests of detainees.

At the same time, the elaboration of a list of non-derogable minimum safeguards can also assist judicial institutions in the prosecution of members of NSAGs who either deny or are indifferent to their obligations under IHL. The more certain we are about the applicable legal framework, the easier it will be to determine whether a particular act constitutes a violation of IHL applicable to NIAC.

4.4. Complementing IHL Standards by the IHRL Provisions

In the absence of clear instructions for detaining authorities to follow in cases of detention in the situation of NIAC there is a temptation to fill in ‘lacunas’ of IHL with the IHRL provisions. This is quite understandable as IHRL contains significantly more elaborate protection for those detained. These provisions are also supported by decades of practice of human rights bodies and a great amount of case law. Continued application of IHRL instruments during NIAC (to the extent they are not derogated from) is no longer questioned either in doctrine or in the case law.Footnote ¹⁷⁶ Therefore, when we are considering detention carried out by states, especially if it occurs far from active hostilities or for reasons not related to armed conflict, IHRL provisions can provide useful means of interpretation of due process guarantees both for internment and for detention in the course of criminal prosecution.

The situation is far less clear with NSAGs, which are typically not considered as holders of human rights obligations. Thus, it is questionable whether provisions of IHRL regulating administrative detention can potentially complement the standards embodied by IHL as applicable to NSAGs and, if so, to what extent. The determination of the applicability of IHRL in relation to internment administered by NSAGs will inevitably affect the scope of procedural guarantees available to the detainees and corresponding obligations of the detaining authorities. Therefore, this article will proceed with an analysis of whether customary international law has developed to the extent of accepting NSAGs as holders of separate obligations under IHRL or whether IHRL is to remain as a subsidiary method of interpreting standards that are already incorporated into IHL and described in Section 4.3.

It is true that, unlike IHL, IHRL does not put states on an equal footing with NSAGs, being created originally to regulate the vertical relationship between the obligation holders (states) and the rights holders (individuals).Footnote ¹⁷⁷ This area of law does not explicitly provide NSAGs with any (even limited) personality, while violations of its provisions do not normally lead to individual accountability of its members based solely on provisions of international law.Footnote ¹⁷⁸

Nevertheless, the reality on the ground is changing. Nowadays we are dealing with situations in which groups like Hezbollah, Taliban, ISIS, Al-Qaeda and Hamas – while participating in hostilities and retaining their status as NSAGs – maintain public order, adopt legislation and establish a system of courts, the jurisdiction of which goes far beyond violations of IHL.Footnote ¹⁷⁹ The need to regulate this practice in order to ensure proper protection of the population living under the control of such groups has led to some scholars insisting on the applicability of IHRL provisions to the conduct of NSAGs. The existence of human rights obligations of NSAGs is supported, in particular, by Jean-Marie Henckaerts and Cornelius Wiesner,Footnote ¹⁸⁰ Andrew Clapham,Footnote ¹⁸¹ Lawrence Hill-Cawthorne,Footnote ¹⁸² and Aristotelis Constantinides.Footnote ¹⁸³ However, even among scholars who support the existence of human rights obligations of non-state actors, there is no unified position with regard to the following:

• • The origins of human rights obligations of NSAGs. The application of IHRL provisions at different times was justified by the exercise of territorial control and quasi-government functions,Footnote ¹⁸⁴ the direct creation of responsibilities as a result of limited legal personality provided by IHL,Footnote ¹⁸⁵ and the customary nature of obligations.Footnote ¹⁸⁶

• • The scope of human rights obligations of NSAGs. The commentators cannot agree on whether NSAGs are to be held responsible for ensuring the whole spectrum of human rights depending on their capacitiesFootnote ¹⁸⁷ or are entrusted only with protection of some key rights already envisaged by IHL.Footnote ¹⁸⁸

• • Types of NSAG – holders of human rights obligations. While some scholars argue that IHRL applies only to NSAGs that exercise territorial control and quasi-governmental functions,Footnote ¹⁸⁹ others are ready to consider all NSAGs as holders of obligations under this area of law.Footnote ¹⁹⁰

It is difficult to imagine that the applicability of IHRL to NSAGs represents lex lata (the law as it exists) when even the proponents of such a conclusion are uncertain about the holders of the obligations and the scope of those obligations. Furthermore, if one accepts that only a limited number of groups that are characterised by a sophisticated hierarchy and administration are obligated by relevant IHRL norms specifying only key rights ‘duplicated’ in IHL, then the complementarity of provisions of IHRL is unlikely to give the international community any practical advantages for the protection of detainees.

Additional risks might come with the artificial differentiation of the scope of obligations of various NSAGs. If an additional criterion, such as the exercise of quasi-governmental functions, is introduced with the aim of imposing additional obligations on a particular type of NSAG, it might lead to a reverse effect of the deliberate unwillingness of potentially qualifying groups to perform these functions in order to avoid the additional burden of responsibility. Such abstention from performing quasi-governmental functions would not necessarily benefit the civilian population living in territories controlled by NSAGs.

Therefore, it would be premature to hold NSAGs fully accountable for compliance with legal standards as developed in the practice of relevant human rights bodies. The difference in standards and procedural guarantees available to detainees captured by states and those who have fallen into the hands of NSAGs might be not satisfying, however this is to be expected. The general inequality of states parties to NIAC outside the scope of IHL has long been established in public international law (see Section 2.3).

It is clear that the practice is at least moving in the direction of forming the basis for establishing certain human rights obligations of NSAGs at the level of custom. References to human rights obligations of NSAGs could be found in the resolutions of various UN bodies including, in particular, the Human Rights CouncilFootnote ¹⁹¹ and the UNSC.Footnote ¹⁹² However, at present both the Human Rights Council and the UNSC in their resolutions refer to the most fundamental human rights that are already incorporated in the limited protection provided by IHL. Therefore, the task of defining the scope of obligations of NSAGs in respect of the treatment of detainees inevitably leads us to the guarantees and safeguards that have their legal basis in conventional or customary IHL.

Certainly, IHL provisions do not exist in a legal vacuum. Thus, the scarcity of conventional rules specifying the content of concrete procedural guarantees provided to internees during NIAC can be remedied (at least to a certain extent) by relevant legal standards developed in IHRL. Ultimately, it would not imply the imposition of additional obligations compared with AP II. Rather, the complementarity in the interrelationship between these areas of law would assist in a meaningful interpretation of current obligations. The extensive practice of states in implementing human rights provisions defining the content of similar rights and safeguards, as well as the jurisprudence of human rights bodies, lead to further development and specification of procedural guarantees and should inform our reading of relevant obligations codified in customary IHL. The growing acceptance of NSAGs being liable for violations of human rights makes it even more justified.