This article is written by James Ugwoke. This article examines state responsibility in proxy warfare, focusing on the effectiveness of the effective control test in attributing the wrongful acts of non-state armed groups to sponsoring states. It explores the accountability gaps in international law and distinguishes between responsibility for proxy conduct and a state’s independent responsibility for its own unlawful actions.
Introduction
Proxy warfare is a military conflict in which one or more third parties directly or indirectly support one or more state or non-state combatants in an effort to influence the conflict’s outcome and thereby to advance their own strategic interests or to undermine those of their opponents. It is a curious beast. If a state wants to punch its rival but would rather not leave fingerprints, it finds a friend: a militia, an insurgent group, or an armed movement with grievances and guns and lets that friend do the punching.
In contrast to traditional wars between states, where opposing nations directly engage each other with their official military forces, proxy warfare relies on the indirect involvement of non-state armed groups by governments to achieve strategic military aims while evading the legal and political repercussions linked to direct military actions. The preference of proxy warfare by some states is due to its capacity to offer sponsoring nations credible deniability, enabling them to extend military influence beyond their borders. By utilizing non-state armed groups instead of conventional military units, countries can often escape accountability for internationally wrongful actions and diminish the political repercussions of military involvement.
This creates strategic advantages for states but also creates a critical legal problem: who bears responsibility when the proxy commits an internationally wrongful act? This problem isn’t cut-and-dry because the mere existence of a relationship between a state and an armed group does not necessarily make the state responsible for every act committed by that. Thus, this paper examines the adequacy of the effective control test as the prevailing standard for attributing the conduct of non-state armed groups to states engaged in proxy warfare. It adopts a doctrinal legal research methodology, examining treaties, customary international law, ICJ judgments, ICTY jurisprudence, ILC materials, academic literature, and selected contemporary examples.
The central research question is straightforward: to what extent does the effective control test adequately determine when a state is internationally responsible for the conduct of a non-state armed group acting as its proxy? The thesis advanced here is that the effective control test remains the prevailing standard for attributing the conduct of non-state armed groups to states, but its demanding operation can create an accountability gap in modern proxy warfare, where states may exercise substantial influence through financing, weapons, intelligence, training, logistics, and strategic coordination without directing the specific operation in which the wrongful act occurs.
The paper does not argue that the effective-control test should simply be abolished. A more sophisticated argument is that the law needs to distinguish between attributing the proxy’s conduct to the State under Article 8, and holding the sponsoring State independently responsible for its own conduct, including unlawful intervention, use of force, aid or assistance, or failure to ensure respect for international humanitarian law.
The Law of State Responsibility
The principle of state responsibility constitutes one of the foundational doctrines of public international law. Under international law, states incur responsibility where conduct attributable to them constitutes a breach of an international legal obligation. However, to attribute the conduct of non-state armed groups to states presents legal difficulties, as these groups occasionally maintain organizational autonomy with distinct command structures and political objectives despite foreign assistance.
The International Law Commission’s Articles on Responsibility of States for Internationally Wrongful Acts (ARSIWA) provide the principal legal framework governing attribution. Under Article 2, an internationally wrongful act exists where conduct is attributable to the state under international law, and the conduct constitutes a breach of an international obligation binding upon that state. Article 4 addresses the conduct of State organs, but proxy forces present a problem because a proxy is generally not formally a state organ. Therefore, one moves to Article 8, which provides that the conduct of private persons or groups may be attributable to a state where they act on the instructions of, or under the direction or control of, that state. Although this provision appears straightforward on paper, its practical application has generated considerable judicial disagreement regarding the degree of control necessary before attribution may arise.
Article 8 and the Attribution Problem
Article 8 recognizes three independent bases of attribution: acting on the instructions of a state, acting under the direction of a state, and acting under the control of a state. The ILC commentary defines “instructions” to include ordering an attack, directing a specific operation, specifying a target, or commanding a particular military action. “Direction” is broader and potentially more difficult. “Control” is the controversial element, and it leads to the most important cases, which are discussed below.
One point needs to be made clear. The phrase ‘effective control’ does not appear directly in Article 8; rather, it emerged from judicial interpretation, especially in the ICJ’s decision in the Nicaragua case, and was later confirmed in Bosnia v. Serbia. The paper therefore looks not at the exact words of Article 8 but at how international courts have interpreted the customary rule that the article embodies.
Nicaragua v. United States
In the Military and Paramilitary Activities in and against Nicaragua v. United States case, the International Court of Justice (ICJ) held that attribution requires proof that the sponsoring state exercised “effective control” over the specific military operations giving rise to the alleged violations. Thus, concluding that although the United States had financed, trained, equipped, and organized the Contra rebels, it was insufficient to attribute all of the rebels’ conduct to the United States because effective operational control over each specific operation had not been established. The court essentially distinguished the state’s own wrongful conduct from wrongful conduct committed by the proxy. The United States could be responsible for its own unlawful conduct without necessarily being legally responsible for every violation committed by the Contras.
Is this threshold too demanding for modern proxy warfare? A lower threshold could mean that virtually any state providing weapons or funding becomes responsible for everything the recipient group does. That would create serious problems of fairness and legal certainty. This tension is exactly what this paper explores.
Tadić and the Overall Control Test
In contrast, the International Criminal Tribunal for the Former Yugoslavia (ICTY), in Prosecutor v Tadić, adopted the less restrictive “overall control” test. The Tribunal reasoned that where a state organizes, coordinates, equips, finances, and generally supervises an organized armed group, this broader relationship may suffice to internationalize the conflict and establish responsibility without requiring operational control over every individual military action. The ICTY adopted the approach because modern proxy warfare frequently involves organized military structures whose operations are shaped by long-term strategic direction rather than day-to-day operational commands.
Bosnia v. Serbia
The ICJ revisited the issue in the 2007 Application of the Genocide Convention judgment (Bosn. & Herz. v. Serb. & Montenegro, 2007). The Court expressly considered the Tadić approach and said it could not accept the overall-control approach for determining state responsibility. It maintained that effective control must be demonstrated in relation to the operations during which the wrongful acts occurred. The ICJ was concerned that overall control would expand state responsibility too far. Its fundamental concern was that a state should be responsible for its own conduct, and attribution requires a sufficiently close connection between the state and the conduct in question. This gives a fantastic analytical tension: ICJ effective control equals stronger connection, narrower attribution, greater legal certainty, but potentially weaker accountability. ICTY overall control equals broader relationship, potentially broader attribution, and potentially greater accountability, but there is a risk of extending state responsibility too far.
Is Effective Control Still Adequate?
There are arguments supporting effective control. Legal certainty: States should not become responsible for every action of a group they merely support. Causation and attribution: There should be a sufficiently close connection between state conduct and the wrongful act. Preventing unlimited responsibility: Otherwise, if a state gives weapons and the armed group commits an unrelated crime, then it automatically becomes responsible.
But there are also arguments against the strict application of effective control. Modern proxy warfare is indirect by design. States may deliberately structure relationships to avoid direct operational control. Plausible deniability: The entire strategic value of a proxy can be that the sponsor does not openly command individual operations. Technological warfare: Control can occur through intelligence, surveillance, satellite information, targeting assistance, communications, cyber capabilities, and sophisticated weapons systems. Evidence problems: The state that controls the relevant information may have every incentive to conceal it. Accountability gap: A state could exercise substantial Convention influence over an armed group while ensuring that it never gives an instruction for that can be directly linked to the specific wrongful operation. That creates a potentially serious gap between practical influence and legal attribution.
Complementary Routes to Responsibility
There is a temptation to frame everything as: “Can we lower the effective-control threshold?” But what if attribution is not the only route to State responsibility? A State may not be responsible for the proxy’s act, but it may nevertheless incur responsibility for its own conduct.
Article 16 of the ARSIWA concerns a State that aids or assists another State in the commission of an internationally wrongful act. However, the ILC commentary makes it clear that Article 16 concerns assistance by one State to another State. Therefore, one should not incorrectly use Article 16 as though it automatically applies to every State-proxy relationship. The wording is about assistance to another State. One should therefore discuss Article 16 carefully, while considering whether other doctrines or specific IHL rules provide better routes in proxy situations.
Common Article 1 of the Geneva Conventions requires States to respect and ensure respect for the Geneva Conventions in all circumstances.
The ICRC’s customary IHL materials treat the obligation to ensure respect as requiring States not to encourage IHL violations and to use their influence, to the degree possible, to stop violations. This is extremely interesting for proxy warfare. Because Article 8 asks: Did the proxy act under the State’s instructions, direction or control? Whereas Common Article 1 raises a different question:
What obligations does the supporting State have to ensure respect for IHL by actors it influences? That distinction could potentially help address the accountability gap without simply abandoning effective control.
The statute concerning the use of force and non-intervention provides a third option: a State
which gives substantial support to armed attacks against another State
can be held responsible for its own illegal actions
even if the specific actions of the proxy are not attributable to it.
The issue is that these routes do not take the place of attribution; rather, they serve as complements. The effective control test should still be the proper criterion for deciding.
when a proxy’s actions count as the State’s actions. It must not be the sole standard by which international law evaluates what the sponsoring State has done.
This response also deals with the argument that stacking
these routes amounts to overall control achieved by means of the back door. It does not. Overall control broadens the scope of attribution: more of the proxy’s actions become the actions of the State. The complementary routes keep attribution narrow and judge the State’s own conduct according to its own standards. The proxy’s war crime stays the proxy’s war crime. But the State’s decision to provide it with arms, funding,
and to coordinate the wider campaign may constitute its own wrong, to be assessed on its own account.
Contemporary Illustration
The Houthis and Hezbollah provide useful illustrations. Numerous United Nations reports have documented varying forms of Iranian support for the Houthis, including
financial assistance, transfers of ballistic missile and drone technology, military training, intelligence cooperation, weapons smuggling
, technical assistance in missile assembly, and political support. However, the Houthis possess an indigenous political leadership, their own command hierarchy,
territorial administration within Yemen, independent recruitment structures, domestic taxation systems, local intelligence networks,
and autonomous strategic objectives connected to Yemeni politics.
These characteristics suggest that the Houthis operate as
an organized non-State armed group rather than as an organ of the Iranian State .
Applying the ICJ’s effective control standard requires considerably more than proof of financial assistance or weapons transfers. The available public evidence demonstrates substantial military cooperation between Iran and the Houthis. However, evidence establishing that Iranian officials directed particular Houthi operations remains limited in the public domain. This examples illustrate the continuing gap between evidence of extensive support and
the legal threshold required for attribution under the effective control test.
Reform and Proposed Approach
When faced with the accountability gap, there is a tendency to opt for the straightforward solution: to lower the threshold. Instead of holding a state responsible only for actions it takes, one would hold it responsible for all actions.
This is a mistaken approach. The fact that the International Court of Justice rejected in the case of Bosnia v. Serbia the use of the overall control standard in relation to state responsibility shows that
there is a genuine principle at stake
namely that a state should be answerable for its own actions and not for all the actions
of any group that it happens to support.
If the threshold for attribution is lowered too much, accountability is lost and what results is a rule
according to which any state that arms any group will be held responsible for everything that group does. The only states that would thereby escape liability are those which are too weak to be significant.
The control test should therefore still be carried out, but it must no longer be regarded
as the only question that matters.
A mistake made in a great deal of the literature is to regard attribution
under Article 8 as the starting point and the final stage of state responsibility in the context of proxy warfare. Where the actions of the proxy are not attributable, the state which is sponsoring them escapes responsibility. This way of putting it produces the gap. It is a misrepresentation. The issue dealt with by attribution is merely this: has this action been carried out by the State? It does not deal with the other question, namely whether the State itself has done something unlawful.
They are distinct inquiries, and international law already has available to it the means needed for the second case. Article 16 is an example of this, with the exception already mentioned. Common Article 1 is another, since it places obligations on States regardless of whether or not attribution has been made. The rules concerning the use of force and the principle of non-intervention provide a third option. All of these deal with the own behaviour of the sponsoring State rather than that of the proxy. They can be breached even if attribution has not been established.
The principle of effective control is still the appropriate criterion for deciding
when a proxy’s actions amount to the actions of the State.
It must not be the only criterion by which international law assesses what the
state which has sponsored an action has done. This reform is more moderate than replacing the rule of effective control and is also the more truthful one.
International law has no need to feign the idea that proxies are state organs
if it is to hold the sponsoring states to account;
all it needs to do is to stop treating attribution as the only question that merits consideration.
Conclusion
The effective control test is still the standard used by the International Court of Justice
when assigning the actions of non-state groups to States and this approach is sound from a
doctrinal point of view and should therefore be kept.
However, the high threshold set by this test can make it difficult to establish attribution in cases
where a State gives extensive support without there being evidence to show the necessary link to the specific wrongful operation. This is what is known as the accountability gap.
The solution is not to substitute effective control for overall control since the ICJ,
has reasonably objected to such a substitution and there are indeed real risks in extending State responsibility too far.
What should be done is to acknowledge that the question of attribution is just one of several issues and that ,
international law should also look into whether the actions of the sponsoring State violate some independent international obligation. The test of effective control is by no means out of date; it is merely incomplete.
