GC IV: Internment in Non-Occupied Territory

International humanitarian law (IHL) treaties use a range of terms to denote deprivation of liberty, one of which is internment. While there is no legal definition, internment is understood to mean the non-criminal detention of a person based on the serious security threat his or her activity poses to a detaining authority. It is the most common form of deprivation of liberty in armed conflict, especially international. This contribution to the joint EJIL Talk /Just Security/ICRC blog series on the occasion of the online publication of the ICRC’s updated Commentary on the 1949 Fourth Geneva (Civilian) Convention first provides a brief reminder of the concept of internment. It then focuses on a particular interpretive elaboration related to the application of IHL obligations in non-occupied territory, highlighting the issue of internment. It ends with a few final remarks.

The Basics of Internment in IAC

Large parts of two Geneva Conventions of 1949 applicable in international armed conflict (IAC) deal with internment.  Under the Third Geneva Convention (GC III) prisoners of war (POWs) are members of the armed forces of a Party to an IAC who have fallen into the power of the enemy. Their internment begins immediately and may last until the end of active hostilities. No process for assessing the lawfulness of POW internment is provided for in the Third Convention because POWs are for the most part captured combatants who have a right to directly participate in hostilities and thus constitute a security threat based on their status. They are not only likely to go back to the fight if released, but in many cases have a duty under the relevant domestic law to attempt to rejoin their forces even if interned.

While prisoner of war internment is relatively straightforward, civilian internment, dealt with in the Fourth Geneva Convention (GC IV), is a much more sensitive issue both practically and legally. Civilians do not wear uniforms and, in reality, there are many different scenarios in which they may be deprived of liberty, such as in a combat zone, a house search, or at a checkpoint. Under GC IV, a protected person is any person who is not eligible for POW status and finds him or herself in the hands of an adverse State Party to the conflict of which they are not nationals. While, as already mentioned, members of the armed forces expressly have a right to participate directly in hostilities civilians do not. Civilian direct participation is not a war crime, but the consequences of such conduct may be dire, including targetability and internment for security reasons.

The “architecture” of GC IV regarding internment starts with Article 27, applicable to protected persons in both the home territories of the Parties to an IAC and in occupied territory. It is a bedrock provision which inter alia requires humane treatment, prohibits adverse distinction, highlights the situation of women (however unsatisfactorily from today’s point of view), and lays out a number of prescriptions and proscriptions comprising the Convention’s basic protective envelope. The last paragraph of the Article reads: “However, the Parties to the Conflict may take such measures of control and security (emphasis added) in regard to protected persons as may be necessary as a result of the war”. This paragraph differs from the rest of the text in terms of content, but its placement is nevertheless telling. The drafters clearly intended to ensure that any measure of control and security applied to civilians will be guided by the all the safeguards of Article 27.  More than sixty articles of GC IV (79-141) subsequently regulate in great detail the specific treatment and conditions of detention required for civilian internees. The remainder of this contribution is focused mainly on procedural aspects.

The Fourth Convention expressly provides a legal basis for civilian internment in a Party’s own territory, as well as in occupied territory, by specifying the grounds and process for internment. In either context, it is clear that this form of deprivation of liberty should be exceptional. In its own territory a State may order internment on the ground that its security “makes it absolutely necessary”, while an Occupying Power may do so if it considers such a measure “necessary, for imperative reasons of security”. GC IV does not provide the specific factual reasons for which conduct may be deemed to reach the requisite level of security threat and in reality no such enumeration could be drawn up.

There is no doubt that direct participation in hostilities by a civilian would meet the threshold of seriousness, but other acts not reaching the level of direct participation may as well (operational intelligence gathering, the specific financing of military operations, training for specific operations, etc.). The burden is on the detaining State to determine what conduct meets the high bar for internment. This should be done in good faith, having also in mind that unnecessary detention is not operationally useful, drains resources, saps staff morale and can affect perceptions among the population, especially abroad, about the legitimacy of a military campaign as a whole.

The Fourth Convention also provides rules on the procedural safeguards that must be applied, in each individual case, to determine whether a civilian represents a security threat warranting internment. The formulation of the relevant rules differs for a State’s own territory and occupied territory. The broad gist of both is that after an order or decision on internment is taken by the Detaining/Occupying Power, an internee has the right to have such a determination promptly reviewed by a competent body/in a regular procedure – by an administrative board or court – and thereafter every six months. The purpose of the initial, and subsequent review(s) is to establish whether the reasons which necessitated internment still exist. If that is not the case, an internee must be released. Civilian internment must in any case end after the close of military operations in an IAC.

States will in practice need to further elaborate the grounds and process for internment given the relative paucity of the relevant provisions mentioned above. Regrettably, this type of regulation tends to be lacking or is generally inadequate even where certain States’ military manuals include it.

GC IV Obligations in Non-Occupied Territory?  

It will remain a mystery, at least to this author, why exactly the drafters of GC IV adopted a treaty text and structure that fail to expressly regulate the invasion phase of an IAC given the background of World War II against which it was negotiated. GC IV protects the “whole of the populations of the countries in conflict” in a small albeit important number of opening articles, while its major part deals with civilian protection either in the home territories of the Parties to a conflict or in occupied territory, or both. This normative layout has generated much expert and scholarly writing and is particularly relevant for the issue of internment.

According to some views, based on the express structure and wording of the text, internment grounds and procedures, as well as the Convention’s more than sixty specific rules on treatment and conditions of civilian internment mentioned above, are not applicable in the invasion phase of an IAC preceding occupation (see Chapter 1, here). Under this strand of argument, this is not what the drafters intended, and the implementation of internment protections would not be feasible in an invasion, which is generally a dynamic and fast-moving operational situation. It is also said that protection gaps should be filled by recourse to customary IHL and applicable human rights law (see Chapter 2, here).

In contrast, it may be argued that a strict delineation between the IAC phases of invasion and occupation does not always correspond to reality. The invasion phase of an IAC may be prolonged, resulting in the non-application of internment protections for detained civilians. It may likewise not be possible to clearly distinguish between the end of an invasion and the beginning of occupation, with the same effect on detained civilians as just mentioned.  

Most important, perhaps, is that an invasion may not necessarily lead to the establishment of an occupation as per the widely accepted definition of the 1907 Hague Regulations (article 42): “Territory is considered occupied when it is actually placed under the authority of the hostile army. The occupation extends only to the territory where such authority has been established and can be exercised”. An example is the United States and other countries’ capture and holding of detainees in the IAC in Afghanistan between October 2001 and June 2002, which was not deemed to have become an occupation and was then reclassified as a NIAC. Insistence, in similar scenarios, on the establishment of a stable 1907 Hague Regulations-like occupation before procedural internee protections are triggered would obviously deprive detainees of safeguards essential to protecting their liberty, health and well-being. In this context it should be borne in mind that customary IHL law, while hugely important, is of a general nature as regards procedural protections and that certain States are known not to accept the extraterritorial application of human rights law.  

The rigid distinction between invasion and occupation has over time been subject to more flexible interpretations that may serve to fill the GC IV protection gap. Already in 1958 Jean Pictet, the author of the ICRC’s first Commentary on GC IV, stated that the word “occupation” as used in GC IV has a wider meaning than in the 1907 Hague Regulations.  He observed that “There is no intermediate period between what might be termed the invasion phase and the inauguration of a stable regime of occupation. Even a patrol which penetrates into enemy territory without any intention of staying there must respect the Conventions in its dealings with the civilians it meets”. According to Pictet, the relevant criterion for application of the Convention’s provisions is thus control by enemy forces over a person, rather than over territory. Pictet’s view may have laid the groundwork for similar interpretations that followed, described below, but regrettably failed to gain traction even among IHL experts until fairly recently. 

The functional approach to occupation

The concept of “functional” occupation was espoused by the ICRC in its 2015 “IHL Challenges Report: “The ICRC considers, however, that in some specific and rather exceptional cases – in particular when foreign forces withdraw from occupied territory (or parts thereof) but retain key elements of authority or other important governmental functions usually performed by an occupying power – the law of occupation may continue to apply within the territorial and functional limits of such competences. Indeed, despite the lack of the physical presence of foreign forces in the territory concerned, the retained authority may amount to effective control for the purposes of the law of occupation and entail the continued application of the relevant provisions of this body of norms. This is referred to as the “functional approach” to the application of occupation law. This test will apply to the extent that the foreign forces still exercise, within all or part of the territory, governmental functions acquired when the occupation was undoubtedly established and ongoing.” (Page 12)

Scholars have likewise taken up the concept of functional occupation, a term first used, to give credit where it is due, by Professor Aeyal Gross. More recently, Professor Marco Sassoli has pointed out that if the IHL of military occupation does not apply in the invasion phase, invading forces would arguably have no legal basis to arrest and detain civilians who threaten their security. He too espouses a functional approach, but has specified that Pictet went “too far in suggesting that control over a person in a territory which is not the invader’s must be sufficient to trigger the application of GC IV to that particular person”. In Sassoli’s view, determining which rules apply in the invasion phase should not be based on “pre-established broad categories”, but on a “sliding-scale” approach that analyzes in each instance whether a particular rule is applicable depending on the degree of control an invader exercises in a given case (paras 8216-8226).

The International Court of Justice (ICJ) also contributed to the functional approach (while not calling it that), in its non-binding 2024 Advisory Opinion on Palestine (paras 86-94), and in its 2025 Advisory Opinion on UNRWA (paras 85-87). In the former the Court, inter alia, opined on whether and how Israel’s withdrawal of its physical military presence from Gaza in 2005 affected its obligations under the law of occupation in that area. The ICJ reiterated the validity of the 1907 Hague Regulations definition, but stated in para 92: “Where an occupying Power, having previously established its authority in the occupied territory, later withdraws its physical presence in part or in whole, it may still bear obligations under the law of occupation to the extent that it remains capable of exercising, and continues to exercise, elements of its authority in place of the local government”. It concluded that Israel’s obligations have remained “commensurate with the degree of its effective control” over the Gaza Strip.

As is obvious, the ICJ dealt with a scenario in which the legal ramifications of the possible end of an occupation were at issue and not whether an occupation was established after an invasion. It is submitted that there would be no cogent practical reasons not to apply the same “commensurate” criterion to other possible obligations provided for in GC IV in the initial phases of an IAC, regardless of whether the fighting has led to a “stable” occupation.

The updated GC IV Commentary

The updated Commentary on GC IV revisits the issue of whether some occupation rules may be relevant in the invasion phase. The commentary on Article 6(2) of GC IV (“Beginning and End of Application of the Convention”) reads: “In the ICRC’s view, there are good reasons to apply some occupation rules during the invasion phase. Occupation law rules have, in general, been calibrated in relation to a certain level of control that can only be observed when the territory is considered occupied for the purposes of international humanitarian law. However, not all rules applicable in occupation necessitate effective control over territory or a well-established and consolidated government-like administration run by the occupying forces in order to be implemented. Governance-related norms and the occupant’s positive obligations generally depend on effective control of a territory. By contrast, individual-related rights and the occupant’s negative obligations do not necessarily require effective territorial control to be implemented. Negative obligations and individual-related rights can apply immediately during the invasion phase. Positive obligations and governance-related norms, on the other hand, would take effect gradually with the progressive solidification of foreign forces’ control over the invaded territory” (footnotes omitted).

By non-exhaustively listing certain articles reflective of “individual-related rights” in a footnote to the above paragraph, the Commentary addresses a possibly justifiable critique of the essentially conceptual nature and thus vagueness of the functional approach. For the purposes of this discussion it is important that GC IV article 78 outlining the grounds and procedure for internment in occupied territory is included. (By way of reminder these are “imperative reasons of security” and a “regular procedure” with the “right of appeal” and periodic review, respectively.) This implies that States detaining civilians in the initial, invasion phase of an IAC may not simply leave them in places of detention until a stable regime of occupation is established (if ever), but should plan detention operations so as to respect their right to be protected by the procedural guarantees of GC IV. The need for such protection is evident: to prevent or put a stop to possibly unlawful confinement – to use the IHL term – which, it should be recalled, may amount to a grave breach of GC IV. In case a recent real-life example is called for one need only think of the arbitrary and abusive detention of civilians reported at the beginning of Russia’s invasion of eastern Ukraine. No GC IV procedural or treatment safeguards specific to civilian internment are known to have been implemented then, or now.

Final remarks

There is a gap in the protection of civilians interned in non-occupied territory. GC IV is silent as regards the invasion phase of an IAC and its structure allows for the exclusion of a detaining State’s procedural and specific treatment obligations for civilian internees in non-occupied territory. The rigid distinction between occupied and invaded/non-occupied territory seems, however, to be eroding due to the increased acceptance of a “functional” approach to occupation by experts, scholars, and the ICJ, whether that term is actually used or not.  This interpretive approach may be contested as a matter of binding law, but should not and cannot be contested by anyone even vaguely familiar with the reality of internment in non-occupied territory.  It is to be hoped that the functional approach will be further discussed and elaborated, and eventually accepted by States, as the ultimate bearers of IHL obligations in IAC.  

Note: This post forms part of a joint symposium with the International Committee of the Red Cross (ICRC) and the editors of Just Security, sharing expert contributions on selected topics addressed in the updated ICRC Commentary on the Fourth Geneva Convention.