About the author(s):
Jelena Aparac
Dr. Jelena Aparac holds a PhD in international public law. Her dissertation focused on “International criminal liability of corporations for international crimes committed in non-international armed conflicts”. She also holds an LLM from the Geneva Academy of International Humanitarian Law and Human Rights. Dr. Aparac is a lecturer in international humanitarian law, human rights, international criminal law and international refugee law. She has given lectures and conferences at academic institutions around the world and has contributed as an independent expert to various intergovernmental discussions on issues related to international peace and security. As a field worker and legal advisor for Médecins sans Frontières, Dr. Aparac worked in conflict areas such as South Sudan, Chad/Darfur, the Democratic Republic of the Congo and the Central African Republic.
Contemporary armed conflicts are increasingly characterised by the involvement of non-State actors operating with varying degrees of State support, blurring the traditional distinction between international and non-international armed conflicts (IAC and NIAC respectively), a distinction that remains central to the application of international humanitarian law (IHL). This is particularly the case when non-State actors operate in NIAC with the support of another State – the sponsor. While often described in policy and strategic terms as “proxy warfare,” this phenomenon remains largely absent from the formal vocabulary of international law, revealing a deeper structural tension within IHL’s binary framework. These developments raise complex questions not only for conflict classification, which is necessary to determine the applicable IHL framework, but also for the attribution of conduct and the establishment of accountability. In this context, Eugénie Duss’ Proxy Warfare et droit international humanitaire arrives at a particularly opportune moment, offering a rigorous analysis of how IHL applies to situations of proxy warfare, where traditional legal categories frequently prove inadequate.
A striking illustration of these challenges is the so-called Wagner Group. Operating through a fluid network of entities across multiple jurisdictions, it appears deliberately structured to complicate legal qualification and accountability. Since 2022, and its visible role in Ukraine, Wagner has undergone reorganisation, including rebranding efforts such as the “Africa Corps.” Yet its networks and personnel continue to operate in evolving configurations. Rather than a single entity, it is better understood as a shifting ecosystem, one that resists clear legal characterisation and tests the limits of existing frameworks.
This evolving and deliberately ambiguous structure raises a fundamental legal question: how should international humanitarian law conceptualise actors such as the Wagner Group? Are they proxies, private military and security companies (PMSCs), or mercenaries and, more importantly, what does this ambiguity reveal about the adequacy of existing legal categories, in particular the notion of “belonging” for the purposes of conflict classification and accountability?
Defining Proxy Warfare
As widely acknowledged, proxy wars are not a new phenomenon. Their contemporary significance lies, however, in the challenges they pose to the foundational categories of IHL. Indeed, the notion of “proxy war” does not constitute a recognised legal term and remains undefined in positive law.
Nevertheless, proxy warfare carries important legal and political implications. It refers to situations in which a State pursues its strategic objectives through the instrumentalisation of armed conflict by acting through intermediaries, using proxies as a method of warfare, as noted by Eugenie Duss (p. 11). For the purposes of her analysis, proxy warfare may thus be understood as the indirect exercise of State power through an intermediary organisation (p. 11).
The book opens by revisiting the doctrinal and jurisprudential debates surrounding the criteria used to classify armed conflicts and attribute conduct. In particular, it contrasts the “overall control” test (ICTY, paras 116-145), employed to classify a conflict as an IAC where a State exercises sufficient control over an armed group, with the “effective control” test (ICJ, paras 396-406), which governs the attribution of conduct for the purposes of establishing State responsibility (pp. 23–31). The coexistence of these distinct thresholds has generated sustained criticism, as it may result in a form of normative incoherence (p. 13): a sponsoring State may become a party to an IAC by exercising overall control over a proxy, yet avoid being held responsible for violations perpetrated by that proxy. At the same time, the proxy, often the sole actor operating on the ground, remains the primary agent of conduct, while the international responsibility of non-State actors continues to be highly contested.
To address this accountability gap, Eugénie Duss proposes refining the notion of “belonging.” More specifically, overall control (and, by extension, belonging) could be conceptualized as a lex specialis derived from IHL. Under this approach, State accountability would be engaged within the IHL framework itself, rather than through the general rules on State responsibility. Such a reconceptualization offers a means of reconciling these two legal regimes while preserving their respective normative logics, Duss argues. However, the complexity and opacity of the phenomenon of the Wagner Group cannot be fully captured through this lens alone.
Defining the Proxy Actors
The book proposes analysing proxy actors through the lens of armed groups. Such groups are inherently tied to the existence of armed conflict: they do not operate independently of it and, absent this context, tend either to dissolve or transform into political entities (p. 80). However, IHL does not provide a legal definition of “armed group” which has been constructed through case law and doctrine (see ICTY, Tadic, para 120). In addition, the notion of the “proxy actor,” has been built on the notion of the armed groups, and likewise lacks a settled legal definition.
Eugénie Duss observes that the traditional notion of an armed group is inadequate when the group functions as a proxy. By definition, such groups are controlled by a sponsoring State, making it difficult, even impossible, to observe them as distinct parties to the conflict. Nonetheless, defining the armed group in the context of proxy warfare remains relevant. First, the group must be sufficiently organized to be able to comply with its obligations under the law of IAC (once the conflict will be internationalized). Second, it must possess the capacity to conduct military operations, acts that can be attributed to the sponsoring State and thereby internationalize the conflict that has been qualified as NIAC. For the purposes of her book, Duss identifies a “proxy armed group” as an entity operating under the overall control of, or in close affiliation with, a sponsoring State.
The Wagner Group; a Proxy Actor?
The Wagner Group’s operations are underpinned by a complex network of illegal and legal commercial structures that facilitate both financing and operational deployment. In several cases, these structures take the form of locally registered companies with ties to Russian entities. For example, Lobaye Invest, registered in the Central African Republic (CAR), has financed and trained the FACA (CAR armed forces) while pursuing control over mineral resources, yet it is owned by the Russian legal entity M Finance (see allegation letter AL CAF 1/2021). Similarly, Sewa Security, owned by a Russian citizen but registered in CAR, and EvroPolis LLC, a Russian company contracted by the Syrian government to guard oil and gas facilities in northern Syria (AL RUS 14/2021), illustrate the use of corporate vehicles to enable Wagner’s operations. These entities demonstrate the Group’s strong links with the Russian Federation: while the Group’s core personnel are predominantly Russian citizens, recruitment frequently extends to Syria, Libya, Belarus, USA, North Korea, and other jurisdictions, reflecting the transnational and hybrid character of Wagner’s forces (AL OTH 8/2023).
This ecosystem composed of variety of entities and nationalities makes it difficult to classify the Wagner Group as an armed group sharing the same nationality as the governmental authorities it is ostensibly fighting against. Indeed, the nationality of its combatants alone may suffice to cast doubt on whether the group can properly be qualified as a proxy armed group. Another relevant element concerns the diversity of their operations, which range from the extraction of natural resources and the trafficking of resources and fighters to direct participation in hostilities. Such a multifaceted operational profile blurs the boundaries between military, economic, and private conduct, complicating its legal characterization, making it difficult to observe it even as a proxy actor.
It is against this background that the next step arises: to assess whether the group operates under the overall control of, or in close affiliation with, a sponsoring State. More precisely, the inquiry is whether it can be considered as “belonging” to that State.
The Wagner Group: Belonging to Sponsor State
It is precisely this notion of “belonging” that proves particularly difficult to apply in the case of the Wagner Group. The fluidity of its structure, the ambiguity of its legal status, and the opacity of its relationship with Russia and contracting States, all contribute to a situation in which existing legal tests struggle to capture the realities of contemporary proxy warfare.
One of the central contributions of Duss’ analysis is the focus on the notion of “belonging.” In traditional IHL, the attribution of conduct to a State depends on tests such as overall control or effective control, which assess the degree of influence a State exercises over an armed group. Yet these tests often fail to capture the operational and structural realities of modern proxies. A proxy actor may function as a quasi-independent entity while simultaneously advancing the objectives of a sponsoring State. This however, would not impact on the classification of the conflict as long as the sponsor State exercise the overall control over the proxy group.
The Wagner Group exemplifies this tension. Its personnel, financing, and operations are intertwined with Russian State interests, yet it does not operate as a formal branch of the State. This raises the question: when can an armed group be said to “belong” to a State for the purposes of IHL and State responsibility? Understanding belonging as more than formal hierarchy allows legal analysis to better reflect the lived reality of proxy warfare, where control is exercised through a network of affiliations, contracts, and operational directives rather than formal command structures.
The notion of “belonging” to a sponsoring State can be illuminated by reference to the criteria used to determine prisoner-of-war (POW) status under the law of IAC. Article 4(A)(2) of the Third Geneva Convention (1949) recognizes members of militias and volunteer corps as belonging to a Party to a conflict where they fight on its behalf and that Party accepts both their role and their conduct, provided that certain conditions are met (a command structure, distinctive signs, the open carrying of arms, and compliance with the laws of war). Where these conditions are fulfilled, such individuals may benefit from POW status and are thus primarily viewed as beneficiaries of specific protections, rather than as sources of international obligations.
However, a State may exercise overall control (or even effective control) irrespective of POW status or compliance with the conditions set out in Article 4(A)(2). According to the ICRC Commentary (paras 1004-1009), “belonging” requires both factual engagement on behalf of a Party and acceptance by that Party. In some cases, tacit acceptance may suffice, particularly where a State does not deny that a group is fighting on its behalf.
The Russian State denied the existence of the Wagner Group and its relation to it until 2022. The example of Wagner Group illustrates how intermediaries can enable “plausible deniability,” thereby obscuring the role and responsibilities of the sponsoring State. The question remains, however, whether the Russia or another State (or both) exercised overall control over the group.
Attribution of the Wagner group acts to the sponsor State
Situation in Mali, Central African Republic and Syria
Attribution of Wagner’s acts presents a related challenge. Across CAR, Mali, and Syria, Wagner operates primarily in support of Russian strategic interests, yet it is also motivated by financial gain, concessions, and contracts, criteria that partially resemble mercenary activity without fully satisfying the legal definition. Two additional criteria, nationality and control, are crucial for defining Wagner as a Russian proxy and for clarifying issues of accountability. Unlike classical proxies, Wagner does not operate directly against a target State in these contexts (territorial States where military operations take place); instead, it is recruited and paid by territorial States, effectively serving their interests. Consequently, attributing Wagner’s acts to either the territorial State or Russia is not straightforward. Here, the control criterion becomes central: while it does not determine conflict classification, it is essential for assessing accountability and establishing which State exercised effective control over military operations for the purposes of the attribution of its operations to this State(s).
Situation in Ukraine
In Ukraine, the attribution calculus changes. Russia is both a Party to the conflict and the sponsor of armed groups in eastern Ukraine, situating Wagner closer to the classical proxy framework. While Russian denials and Wagner’s opaque operational structures initially reinforced plausible deniability, 2022 statements by its leader Prigozhin confirmed direct operational engagement. His death further clarified the group’s subordination, heightening the relevance of belonging and attribution for accountability. As in other contexts, Wagner personnel are not nationals of the territorial State (Ukraine), reinforcing that their classification as proxies does not impact conflict qualification. Rather, the notion of belonging primarily informs the attribution of their actions to the Russian Federation for purposes of legal responsibility.
Conclusion
The Wagner Group illustrates the limits of conventional proxy and armed group categories under IHL. Operationally, it may belong to a Party, yet formally remain outside recognised legal categories, highlighting the challenge of assessing the legal implications of proxy actors.
In most contexts, Wagner personnel do not qualify as mercenaries. Article 47 of Additional Protocol I sets cumulative criteria. These criteria, however, rarely fit Wagner: Article 47 applies only to international armed conflicts, direct participation is inconsistent due to commercial and extractive activities, and in cases such as Ukraine, and various nationality excludes them from its scope.
The Wagner Group is also not a PMSC, despite self-identifying as such; PMSCs lack formal recognition under international law, no binding regulation exists, and Russian law prohibits them.
Proxy warfare is increasingly common yet remains poorly defined. States frequently employ proxy groups as instruments of warfare and political strategy, avoiding direct responsibility for their actions. Actors like the Wagner Group blur the lines between traditional legal categories and emerging forms of warfare, leaving both conflict classification and actor attribution legally unsettled.
Belonging and attribution remain intertwined: while operational ties may link a group to a sponsor State, the absence of formal incorporation, the multinational composition of personnel, and financial incentives complicate the application of traditional legal frameworks. Understanding these nuances is critical for assessing both the status of proxy actors and the accountability of States that deploy them.
In this context, Eugénie Duss’s book is a timely and necessary contribution to the field.