GC IV Commentary Series – The Prohibition of Deportation and Forcible Transfer

Editors’ note: The following series considers ICRC’s 2025 updated Commentary on the Fourth Geneva Convention. Rosa-Lena Lauterbach, Liisi Adamson, and Sean Watts’s introduction to the series is available here.

When Russian forces reached a town in Ukraine’s Zaporizhzhia region in 2022, many of its residents filled the streets in pro-Ukrainian rallies. In testimony recorded by The Reckoning Project (para. 55), one woman detailed the fate of those who dared to oppose the invasion. One morning, uniformed men arrived at her home and took her, “handcuffed and with a bag over her head,” to a detention centre. She was kept in inhumane conditions, beaten, sexually abused, and intimidated. On one occasion, her captors threatened her with “an elephant”: “the placement of a gas mask over her face, with soldiers intermittently blocking the air.” The soldiers released her only after forcing her to record a propaganda video for Russia. The day after her release, she took her daughter and fled to a government-controlled area of Ukraine, leaving her home, family, and life behind.

There are countless stories of displacement in times of armed conflict. In the words of an ICRC report, “displacement is part and parcel of war.” In some cases, individuals flee areas of hostilities because of a generalised state of violence and uncertainty. Faced with the risk of harm in their homes, they make the difficult, almost impossible, choice of leaving in the hope of security elsewhere. In others, people are forced to leave against their will, either by being physically taken away or compelled into departure through other means.

Article 49 of Geneva Convention IV (GC IV) is titled “Deportations, transfers, evacuations.” Located in Part III, Section III on occupied territories, it provides the clearest regulation of displacement in international armed conflict under international humanitarian law. In its first paragraph, it prohibits “[i]ndividual or mass forcible transfers, as well as deportations of protected persons from occupied territory to the territory of the Occupying Power or to that of any other country, occupied or not, … regardless of their motive.” While other rules of international humanitarian law, such as the obligations falling within the list of fundamental guarantees, bear on the process and effects of displacement, Article 49 prohibits forcible displacement as a distinct wrong.

Despite its significance, interpretative uncertainties about the applicability and scope of the provision persist. The ICRC’s Updated Commentary on the Fourth Geneva Convention (2025 Commentary) makes an important and timely contribution to the clarification of Article 49. Even a cursory comparison between the Commentary on Geneva Convention IV Relative to the Protection of Civilians in Time of War (1958 Commentary) and the 2025 Commentary demonstrates the detail and depth of the 2025 project. This reflects both the numerous legal developments since the publication of the 1958 Commentary and the monumental work of the ICRC. It consolidates existing interpretations and solidifies and specifies understandings of key elements of the provision. It takes a stand on the contentious question of the temporal application of Article 49. And while, as I argue below, it leaves some points of substance and methodology under-specified, implicit, or dispersed in the Commentary, there is no doubt it has, in a reasoned and substantiated manner, taken the text of the Convention to the realities and demands of 21st century conflicts.

In this post, I will examine the contribution made by the 2025 Commentary by exploring two specific issues: the meaning of “forcible” displacement (displacement is used here as an umbrella term capturing deportation and transfer), and the temporal applicability of Article 49. A sound understanding of these two issues is key to the protection of civilians in contemporary conflicts.

The Meaning of “Forcible” Displacement

Article 49 prohibits “[i]ndividual or mass forcible transfers, as well as deportations of protected persons from occupied territory … regardless of their motive.” In the 1958 Commentary, we read that “the Diplomatic Conference preferred not to place an absolute prohibition on transfers of all kinds, as some might up to a certain point have the consent of those being transferred” and “the Conference decided to authorize voluntary transfers by implication, and only to prohibit “forcible” transfers” (p. 279). The 2025 Commentary, consolidating decades of jurisprudence in the area of international criminal law, clarifies the test for distinguishing displacements that fall within the scope of Article 49 from those that do not: their “forcible” nature, that is, displacements “carried out against the protected person’s free will” (para. 3169).

The “forcible” nature of displacement under the prohibition, alongside the implicit element of intent to transfer on the part of the occupying force (see here, p. 93), play an important limiting role. Without them, the prohibition could end up capturing any voluntary or involuntary decision to leave in situations of occupation, even where displacement was neither a desired nor known outcome of the actions of the occupier.

At the same time, the term “forcible” is not to be read in an unduly restrictive way to only capture physical force involving the physical movement of persons by the occupying power. Of course, direct physical measures to displace, such as the organised deportation or forcible transfer of children (paras. 7-18), and the forcible expulsion of unwanted individuals (paras. 55-72), fall within the purview of the prohibition. However, “forcible” measures or acts that undermine or extinguish the agency and choice of individuals come in a wider variety. As acknowledged by the International Tribunal for the former Yugoslavia in Krnojelac, the term forced may include the “threat of force or coercion, such as that caused by fear of violence, duress, detention, psychological oppression or abuse of power against such person or persons or another person, or by taking advantage of a coercive environment” (Trial Judgment, para. 475). This interpretation of “forcible” brings testimonies such as the one from Zaporizhzhia within the scope of Article 49. Russian forces did not physically carry out the displacement, but their arbitrary detention, torture, and intimidation created or exacerbated a coercive environment, which left the victim with no choice but to leave.

The 2025 Commentary endorses this interpretation of “forcible.” It confirms that forcible transfer or deportation can be indirect. This would be the case “when a party does not expressly order, or is not directly involved in the organization or carrying out of the transfer or deportation but commits violations of international humanitarian law or human rights law with the effect to cause the population to leave, for example by unlawfully destroying houses or imposing collective punishments” (para. 3171). The Commentary also notes that indirect displacement can occur “through the imposition of ‘severe living conditions’ that leave the persons concerned with no genuine choice’’ (para. 3172).

To reach this conclusion, the ICRC relies on the ordinary meaning of the term “forcible” (para. 3170) and the principle of effectiveness in interpretation (para. 3173: “This interpretation gives full effect to the prohibition in Article 49(1). It prevents the Occupying Power from circumventing it by indirectly forcing people to leave without expressly ordering or organizing their departure”). It also relies extensively on the jurisprudence of tribunals and courts for individual criminal responsibility (as acknowledged in para. 122 of the introduction to the 2025 Commentary), which have, over the years, interpreted Article 49 in their analysis of the scope of the war crimes built on its foundations.

The 2025 Commentary often uses such jurisprudence as an argumentative shortcut, i.e. recognising a legal proposition because it has been established in a different legal process (One commentator develops this concept in relation to a distinct methodological process, the identification of custom. Yet the insights are also relevant to shortcuts in treaty interpretation.) Reliance on such jurisprudence is not inapposite. If the judicial decision has itself duly engaged in treaty interpretation, it can especially serve as a valuable subsidiary means for the determination of the provision. (On the interaction between treaty interpretation and subsidiary means for the determination of the law per Article 38(1)(d) of the ICJ Statute, see here.)

Ultimately, the 2025 Commentary consolidates and solidifies a meaning of “forcible” that covers the many ways in which an individual can be deprived of genuine choice. This reading of the obligations finds support from international organisations (see e.g., for instance, here), academia (see e.g., here, p. 181; here, p.73; and here, p. 26) and civil society (see e.g., here). To go back to Zaporizhzhia, the woman’s “choice” to depart following the violation of her human rights and impending threat of further violations by the occupying forces was just as non-existent as that of a person physically forced into a car and transported out of their town.

Beyond this baseline, important questions will remain to be answered on a case-by-case basis. For instance, what are the metrics for the existence of a coercive environment extinguishing genuine choice? Does the coercion need to be direct and/or immediate? The European Court of Human Rights grappled with these questions in Ukraine and The Netherlands v Russia under a claim of unlawful interference with private life per Article 8 of the European Convention on Human Rights. The Court concluded that, in the relevant circumstances at the time, “it cannot be said that those who left their homes in occupied territory to flee the war and violence in Ukraine, either to Russia or elsewhere, including the EU, were doing so of their own free choice even where direct and immediate coercion was not present” (para. 1171). Going forward, it may be that further specification of the meaning of “forcible” will be found in the outputs of human rights bodies interpreting obligations under their constitutive instruments in the light of Article 49 of GC IV as a relevant and applicable rule (see Article. 31(3)(c) of the Vienna Convention on the Law of Treaties (VCLT).

The Temporal Applicability of Article 49

Turning to the second issue, when does Article 49 begin to apply? According to the 2025 Commentary, “[t]he prohibitions in Article 49 are among the rules applicable during occupation that also apply during the so-called ‘invasion phase’” (para. 3162). Whether GC IV rules established under Part III, Section III on occupied territories do indeed apply before territory is considered occupied per the definition of occupation in the Hague Regulations, i.e., “actually placed under the authority of the hostile army” (1907 Hague Regulations, Article 42), has been the subject of debate (see e.g., here, here, and here).

Already in the 1958 Commentary, there was a sense of a protection gap, to be filled through interpretation. Specifically, given its structure, it is possible to construe GC IV as excluding from the scope of Part III, Section III obligations owed to persons under the control of enemy forces outside occupied territory or the territory of a party to the conflict. This has significant repercussions in practice, as people forcibly displaced during an invasion but prior to occupation would fall outside the scope of Article 49. And yet, removing those persons from the area—especially journalists, activists, educators—could be key to establishing effective control over the territory.

However, the 1958 Commentary on Article 6 of GC IV asserts that “[n]o loophole is left.”

the word “occupation”, as used in the Article, has a wider meaning than it has in Article 42 of the Regulations annexed to the Fourth Hague Convention of 1907. So far as individuals are concerned, the application of the Fourth Geneva Convention does not depend upon the existence of a state of occupation within the meaning of the Article 42 referred to above. The relations between the civilian population of a territory and troops advancing into that territory, whether fighting or not, are governed by the present Convention. 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. … The Convention is quite definite on this point: all persons who find themselves in the hands of a Party to the conflict or an Occupying Power of which they are not nationals are protected persons (p. 60).

Thus, according to the 1958 Commentary, the test for determining “occupation” for the purposes of GC IV is different to the one established in the Hague Regulations. This position has generated significant debate. One school of thought challenges the interpretation advanced in the 1958 Commentary for finding no support in a structured interpretation of GC IV, conflating the determination of occupation with the definition of protected persons, and being operationally unrealistic (see e.g., here p. 30-36). By contrast, others have specified the interpretation by suggesting a “functional understanding” of the amount of the territory that a party must occupy, a flexible interpretation of what is owed under different obligations, or, alternatively, rule-specific conceptions of “occupation” (see e.g., here  p. 42-50).

The ICRC itself has, since then, adopted a position that differs from the 1958 Commentary (see here). While it continues to advocate for “a broad interpretation of the application of the Fourth Geneva Convention during both the invasion and withdrawal phases, with a view to maximizing the legal protection conferred on the civilian population” (p. 27), its position is rooted in the extension of specific rules from GC IV to the invasion stage rather than a distinct interpretation of occupation (see also here (p. 7)).

A conjunctive reading of the 2025 Commentary to Article 2 (Application of the Convention), Article 4 (Definition of protected persons), Article 6 (Beginning and end of application of the Convention) and Article 49 elaborates the ICRC position on the applicability of the prohibition of deportation and forcible transfer. First, the meaning of occupation in GC IV does not depart from the definition of occupation in Article 42 of the Hague Regulations of 1907 (see 2025 Commentary, paras. 364, 366, 369). Second, certain rules applicable only during occupation can extend to the invasion phase. According to the ICRC “there are good reasons to apply some occupation rules during the invasion phase” (para. 1140), with a calibration based on the level of control exercised. Thus, “[n]egative 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” (para. 1140). Third, Article 49 is one of the rules whose application extends to the invasion phase. In the commentary to Article 4 we read that “[i]t would certainly run counter to the protective object and purpose of the Convention if, for example, the deportation or forcible transfer of such civilians by the invading forces were not prohibited simply for the reason that such forces have not yet established effective control” (para. 1071).

To understand the legal basis for the ICRC’s position, one needs to collate pieces from the commentaries to various GC IV provisions: the methodological cues are dispersed. At In describing the arguments of proponents of the extension, the ICRC considers that this “would accord with the object and purpose of the Convention,” suggesting an anchoring in treaty interpretation (para. 1138). In contrast, the language of “good reasons to apply some occupation rules during the invasion phase,” and the subsequent discussion of obligations tracking degrees of control do not explicitly refer to any element of treaty interpretation and could be read as a normative argument (para. 1140). Admittedly, footnote 33 to paragraph 1140 sends the reader to a source that discusses, among others, a teleological approach to reach this position, but the source contains a discussion of both legal and policy reasons (see here p. 25). Interestingly, the statement that the prohibitions in Article 49 “are among the rules applicable during occupation that also apply during the so-called ‘invasion phase’” seems much more assertive as a position under lex lata (para. 3162). Overall, it seems that the anchor to this argument is the object and purpose of GC IV. Yet, the 2025 Commentary does not fully, or at least explicitly, explain the legal steps to reach this position.

Methodological questions aside, the position taken in the 2025 Commentary is a significant statement of principle responding to a need for clear positions on civilian protection. An earlier post observed,

[t]he current uncertainty in law and practice on this issue would seem to call for a focused examination by States and other actors with a view to arriving, as a start, at a policy position on the status and treatment of civilians not explicitly covered by GC IV. The normative gap is significant and so, inevitably, is the protection gap.

While the ICRC cannot make or change GC IV, its position may have effects of relevance to its interpretation. It could, for instance, incentivise States to react, generating subsequent practice that either aligns with or departs from the ICRC position, of relevance under Articles 31(3)(b) and 32 of the VCLT.

Even if one were to extend the application of Article 49 to the invasion stage, questions remain. As one commentator notes, “[r]elatively little attention has been paid to the question of when the law of occupation starts to apply, and in particular when an invasion turns into an occupation” (p.30). Even less attention has been paid to the question of the conceptual boundaries of the invasion phase. Does it imply the physical presence of enemy forces in the locality of protected persons? What if individuals decide to leave before these forces reach their towns and villages to avoid the arbitrary detention, torture and other human rights violations that they know will accompany the invasion? Leaving before the arrival of the adversary may, in fact, be the only option to avoid these harms.

One could say that, in these cases, such persons would not qualify as “protected persons” as they do not “find themselves … ‘in the hands of’ a party to the conflict or occupying power” (GC IV, art. 4). As clarified by the ICRC, however, “the expression ‘in the hands of’ need not necessarily be understood in the physical sense of, for example, being detained; it simply means that the person is in territory that is under the control of the Power in question” (2025 Commentary, para. 1037). Though the 2025 Commentary here speaks of territorial control, the provision does not necessarily exclude “control over a person” as a plausible interpretation. It may be that advancing forces can extinguish individuals’ choices at a distance, without coming into any physical contact. Social media and other technologies could exacerbate this concern by becoming easy channels for the exercise of such control. Videos documenting torture, harassment, or intimidation against dissident voices could alert others to the risk of similar treatment. Even at a distance, individuals could be left with no choice but to leave.

Ultimately, as the 2025 Commentary suggests, the non-applicability of specific GC IV obligations does not equal absence of protection. The Commentary situates GC IV within the broader ecosystem of international law. For instance, we read that “[d]espite the continuing debate surrounding the applicability of occupation law rules to the invasion phase, the non-application of these rules by invading forces would not leave the persons concerned without any legal protections under international humanitarian law” (para. 1141). The Commentary then notes the relevance of Part II and Part III, Section I of GC IV, as well as obligations under international human rights law. While it is true that the existence of these rules mitigates the risk of a legal vacuum, it bears emphasis that some of these rules are themselves subject to interpretative contestation (for instance, the meaning of jurisdiction under different human rights treaties), and they do not necessarily or directly address the specific wrongs covered by GC IV rules, the prohibition of deportation and forcible transfer being a case in point.

Conclusion

Legal uncertainties undermine the protection of persons from the horrors of armed conflict. The 2025 Commentary makes an important contribution to the clarification of Article 49 of GC IV. It consolidates interpretations developed over decades, provides much-needed specificity and advances understandings on contentious questions.

It is to be hoped that the contribution of the Commentary will lie not only in its monumental achievement of systematising debates and solidifying interpretations, but also in its generation of further legal clarification by States. This clarity matters to the survivors who have suffered and continue to suffer from displacement. It matters to a woman and her daughter who are missing their home in Zaporizhzhia.

***

Dr Tsvetelina van Benthem is a Lecturer in Law at the University of Reading School of Law and Programme Director of Reading’s Global Law Programmes. Tsvetelina is also a lecturer in international law at the University of Oxford, where she teaches Public International Law for the Diplomatic Studies Programme and International Law and Armed Conflict for the BCL/MJur.

The views expressed are those of the author, and do not necessarily reflect the official position of the United States Military Academy, Department of the Army, or Department of Defense.

Articles of War is a forum for professionals to share opinions and cultivate ideas. Articles of War does not screen articles to fit a particular editorial agenda, nor endorse or advocate material that is published. Authorship does not indicate affiliation with Articles of War, the Lieber Institute, or the United States Military Academy West Point.

Photo credit: Kyiv City State Administration, Olexiy Samsonov