Ukraine Symposium – The Risk of Commercial Actors in Outer Space Drawing States into Armed Conflict

Prior to the Russian invasion of Ukraine, the satellite communications firm Viasat confirmed a distributed denial of service (DDoS) attack that is suspected to have caused a partial outage in its broadband services across KA-SAT, a network that provides internet coverage over much of Europe. The attack impacted Ukrainian government internet websites, banks, and several thousand residential customers. The Five Eyes (United Kingdom, United States, Canada, Australia, and New Zealand) intelligence sharing community attributed the attack to the Russian government. More recently, an attack against Ukrtelecom, Ukraine’s largest telecommunications provider geographically, saw its connectivity collapse to 13% of pre-war levels.

In the midst of these challenges, Elon Musk responded to the Ukrainian government’s tweet requesting that he offer internet services through his constellation of 2,400 SpaceX Starlink satellites. The Starlink satellites are positioned in Low Earth Orbit (LEO) and are launched from inside the United States. Starlink is able to provide a faster internet broadband capability than traditional fibre-optic cables or single satellites operating out of the Geostationary Orbit (GEO). SpaceX made its services available first to the Ukrainian armed forces and emergency services, and then to the wider populace. The service has been highly effective so far, impervious to bombardment and shelling in Mariupol and Chernihiv, as well as jamming attempts by Russia.

However, the provision of Starlink services raises questions about the U.S. role or responsibility in the conflict. This post examines State responsibility for Starlink operations under the Outer Space Treaty (OST), the law of neutrality, and co-belligerency. These issues bear on the legal risk of the United States being drawn into armed conflict by the actions of commercial actors in space.

The Outer Space Treaty and State Responsibility

The United States, Ukraine, and Russia are three of the 110 States that have ratified the OST, with another twenty-three States being signatories to the instrument. In addition to its treaty status, some States believe that the entire OST has customary international law (CIL) status (see, e.g., the German comments during a 2014 session on the Committee on the Peaceful Uses of Outer Space (COPUOS)). Other States only accept “fundamental” rules in the OST as having attained such status.

Under Article VI of the OST, States bear international responsibility for “national activities” in outer space, whether or not such activities are by governmental agencies or by non-governmental entities (NGEs). The term “international responsibility” is widely regarded as synonymous with “State Responsibility” under the Draft Articles on Responsibility of States for Internationally Wrongful Acts (IWAs), which was in the decades-long process of being drafted when the OST was opened for signature. Additionally, under Article VI of the OST, States are responsible for “assuring that national activities are carried out in conformity with the provisions set forth in the present Treaty,” which includes, through application of Article III of the OST, compliance “with international law.” Where activities are carried out by NGEs, Article VI amplifies that States must authorize and continually supervise such activities.

Concerning the concept of international responsibility in article VI of the OST, there are three key matters requiring consideration: (1) State responsibility for national activities, (2) the obligation on States to ensure that national activities conform with the provisions of the OST, and international law more widely and, (3) the obligation on States to ensure that they authorize and continually supervise the activities of NGEs.

(1) The OST’s attribution standard of State Responsibility for national activities

In respect of (1), this is a special attribution rule that differs to that which applies in other domains pursuant to the law of State responsibility. The OST travaux préparatoires provides insight into the negotiations revealing that the Soviet Union opposed any use of outer space by private entities, whereas the United States took a contrary position. Article VI was the compromise reached between the then space superpowers (see Professor Von Der Dunk’s piece for a discussion on “private activity but public responsibility”). In other domains State responsibility would depend upon whether commercial providers were in fact acting on the instructions of, or under the direction or control of, the State (Article 8, Draft Articles on Responsibility of States for IWA’s), which requires consideration of the “effective control” standard set out by the International Court of Justice (ICJ) in Paramilitary Activities and endorsed in the Bosnian Genocide case.

In outer space however, State responsibility depends on whether the activity carried out by an NGE is a “national activity.” The OST does not define the term “national activities,” but State practice demonstrates that it often includes activities from within the State’s territorial jurisdiction or by its nationals (see, e.g., the United Kingdom Space Industry Act 2018), which would include commercial providers of satellite communication). Differing interpretations of what type of activity will constitute a “national activity” for the purpose of State responsibility are summarized in detail by Professor Von Der Dunk and summarized below.

The first interpretation focuses on the use of the word “national” and holds States responsible for all activities conducted by its nationals (including persons and NGEs), irrespective of the level of control that the State may or may not exercise. The second interpretation conflates international responsibility within Article VI with international State liability in Article VII and concludes that “national activities” should refer to activities in which the State was a “launching State.” This interpretation is problematic and does not take account of the different language used in the two clauses. The third interpretation draws on the obligation on States to “authorize and continually supervise” the activities of NGEs and holds States internationally responsible only where they have the tools to control such activities, i.e., where the activities fall within the jurisdiction of the State. These three schools of thought may be applied to the Starlink situation.

The correct view lies somewhere between the first and third interpretations. While States should not be responsible for all activities of their nationals—i.e., the hacking of a satellite by a private individiual—it should be recognized that commercial operators, such as Starlink, could not operate in space without the authorization of the appropriate State and therefore its space activities will constitute “national activities.” This view is consistent with the State practice of regulating commercial satellite providers and meets the intent of the drafters of Article VI to ensure States couldn’t hide behind proxies to avoid State responsibility, as discussed above.

By this view, States are internationally responsible for activities carried out by such NGEs in respect of the State’s obligations under international law, including for IWAs.

(2) The Article VI OST obligation on States to ensure that national activities conform with the provisions of the Treaty and international law more widely

Compliance with the provisions of the OST includes, for example, the obligation in Article IX of the OST to conduct activities with due regard for the corresponding interests of other States. In late December 2021 the activities of Starlink necessitated that China’s space station, on two occasions, manoeuvred to avoid a collision. China called upon the UN Secretary General to remind States of their State responsibility obligations under Article VI of the OST. Notably, China did not assert that the behaviour was in breach of Article IX—probably to avoid its own behaviour being called into question, such as Beijing’s 2007 ASAT demonstration or the more recent failure to reserve enough propellant to allow a controlled rocket re-entry to Earth.

Compliance with the provisions of the OST also includes Article III of the OST, which requires that States carry on activities in accordance with international law more generally, including that of the UN Charter, such as the obligation under Article 2(4) of the Charter prohibiting the threat or use of force.

(3) The Article VI obligation on States to ensure they authorize and continually supervise the activities of NGEs

By ensuring compliance with the more onerous OST attribution, this standard necessitates that States adopt a supervisory role over commercial entities operating in outer space. States, including the United States, Ukraine, and Russia, have implemented domestic legislation relating to the launching and operation of satellites, enabling them to oversee domestic NGEs and ensure conformity with the States’ OST obligations. For example, in the United States, federal licenses are required for launches and re-entry of space vehicles in the United States, and for any U.S. citizens regardless of the site of the launch.

The fact that the United States bears international responsibility for the actions of commercial providers operating from within its jurisdiction under Article VI of the OST means that Musk’s provision of Starlink services to Ukraine is attributable to the United States. This is pertinent to the traditional law of neutrality, under which the United States assumes neutral obligations to abstain from providing material support to belligerent parties.

(Qualified) Neutrality

The section that follows will explain why the provision of Starlink services does not undermine either traditional or qualified neutrality. The law of neutrality is found in two 1907 Hague Conventions—Convention V applies in the land domain, and Convention XIII in the maritime domain. However, the overarching principle of neutrality is applicable to all international armed conflict as a matter of customary law (see paragraph 89 of the Nuclear Weapons Advisory Opinion), including in outer space.

General principles and obligations found in the two Conventions are reflective of their customary counterparts that apply in space. Article VI of Convention XIII prohibits the supply, by a neutral power to a belligerent, of war material. By one view, the provision of military materiel to Ukraine violates obligations under neutrality law, by another it may fall within the parameters of qualified neutrality.

The pertinent questions are whether (1) the provision of access to internet services amounts to the supply of war material; (2) whether the Article VI attribution rule binds the neutral state notwithstanding the freedom of commercial trade practiced under the traditional law of neutrality; and (3) whether the provision of Starlink meets the criteria necessary to bestow qualified neutrality.

As to (1), a total of 15,000 Starlink terminals were delivered to Ukraine and the service has directly impacted military operations. On that basis, the assistance is capable of being equated to military material.

As to (2), there is a telecommunications exemption in Article VIII of Convention V which provides that a neutral power is not required to prevent or restrict the belligerents access to telegraph or telephone cables or wireless telegraphy apparatus belonging to it or to companies or to private individuals. The focus of the language on cables and apparatus is simply due to the passage of time of over a Century since the Convention was entered into. Modern telecommunication systems would also fall within the scope of the exemption, a position that is consistent with the approach in the Tallinn Manual 2.0 (Rule 151) relating to the use of networks for military purposes, which confirms that Article VIII applies to cyber communication systems. As such, the provision of Starlink does not trigger a breach of traditional neutrality.

A short analysis will be given on question (3) as to qualified neutrality as it may be of relevance to future activities by NGEs. Under the DoD Law of War Manual (§ 15.2.2), qualified neutrality permits neutral States to discriminate in favor of States that are victims of aggression, a position consistent with the views of Professor Michael Schmitt in a recent Articles of War post.

Under the United Nations Charter, qualified neutrality would be foreshadowed by a Chapter VII finding of a threat to the peace, breach of the peace, or act of aggression. However, in this case, Russia used its veto power as a permanent member of the United Nations Security Council (UNSC) to block a proposed UNSC Resolution under Chapter VII condemning Russia. In such cases, Schmitt and Wolff Heintschel von Heinegg argue that application of qualified neutrality still makes sense if it is otherwise clear that the State being assisted is the victim of aggression. If this position is accepted, States may provide the victim State with certain assistance, including arms and military equipment, while maintaining a neutral status. (But for a discussion of some protentional limitations of the concept, see Professor Terry Gill’s and Raul (Pete) Pedrozo’s posts.)

How far that assistance may go to favor the victim State without breaching qualified neutrality obligations requires a case-by-case analysis. It is at least arguable that recent intelligence data provided by the United States and NATO allies to Ukraine used to lethally target Russian generals and destroy Russia’s flagship missile cruiser would fall outside the parameters of neutrality law, although not necessarily render the measures unlawful (on intelligence sharing and neutrality see Schmitt and Milanovic). Professor Schmitt concludes that the provision of military material falls on the other side of the spectrum and does not violate qualified neutrality. Simply providing a substitute means of internet access after an aggressor destroys the victim State’s original network is even further removed from the provision of military material, particularly where that service is also available to the civilian population, and does not violate qualified neutrality.

Co-Belligerency

Still, even if Musk’s provision of Starlink to Ukraine were to violate the United States’ neutrality obligations, that of itself does not render the United States a party to the existing international armed conflict (IAC) between Ukraine and Russia.

The criteria for the existence of an IAC is set out in Common Article 2 of the four Geneva Conventions, which is reflective of customary international law. An IAC exists in “all cases of declared war or of any other armed conflict which may arise between two or more of the High Contracting Parties, even if the state of war is not recognized by one of them.”

The International Committee of the Red Cross (ICRC) 2016 commentary to Common Article 2 makes clear that there is no requirement that the use of force reach a certain level of intensity and it would cover even “minor skirmishes” and possibly “any unconsented-to military operations.” The rationale for a low threshold in determining that an IAC exists is to ensure maximum application to those protected under various International Humanitarian Law (IHL) instruments, which govern the conduct of hostilities during an armed conflict. IHL, otherwise referred to as the law of armed conflict (LOAC) sets forth various obligations and protections and its application has ramifications on matters such as prisoner of war status and targetability. For example, Russia could legitimately launch an attack, meaning an “act of violence” (Article 49(1), Additional Protocol I (AP I) to the Geneva Conventions) against U.S. armed forces and military objectives if the United States was party to the conflict.

In light of the international support offered to Ukraine in recent months, there have been numerous observations as to when assistance to a belligerent during an IAC brings the state rendering assistance into the IAC as a co-belligerent. As the financial and arms support to Ukraine by the United States does not violate qualified neutrality as noted above; a fortiori—it does not trigger an IAC between the U.S. and Russia. Likewise, because the provision of Starlink services benefits from the telecommunications exemption, it too does not cause the United States to become a co-belligerent in the conflict. This conclusion also follows when examined under a slightly different analysis.

Article VI attribution is expressly concerned with State responsibility for IWA’s and does not impose responsibility for all purposes, including conflict classification. The International Criminal Tribunal for the Former Yugoslavia in the case of Tadić (a case concerning when a NIAC may become internationalized) expressly adopted a different standard to State responsibility on the question of conflict characterization, relying on a test of “overall control.” As Schmitt notes, when determining whether the provision of support by a state not party to the conflict may cause the non-party to lose that status and join the conflict, “at issue is the degree of attenuation between the support in question and any supportive actions.”

He draws on the ICRC support-based approach as outlined in in its 2019 Challenges report of assessing when a NIAC has become internationalized and acknowledges there is no consensus amongst States as to where the precise boundary lies in judging when support to a belligerent draws the supporting state into a conflict. However, the NIAC analogy is helpful where there is a lack of state practice or jurisprudence in an IAC setting. The mere provision of internet and communication services would not make the United States a party—facilitating the service does not demonstrate intent to contribute to specific conduct of hostility operations, and it is enduring so there is no immediacy between the support and its military use.

As reflected in the exploration of this specific case, the possibility of commercial satellite providers implicating States in a future conflict would depend on the degree of connection between the support and specific conduct of hostility operations and whether the actions of the NGE are under the “overall control” of the State.

Conclusion

The application of Article VI of the OST is limited to the question of State responsibility and does not bear on conflict classification. However, as China’s comment on Starlink being “militarized” demonstrates, while they may not consider the United States as a co-belligerent they are watching the activities of commercial actors in outer space and cognizant that commercial actors may become legitimate military objectives.

Article VI is a unique attribution rule which applies to all IWAs and engages State responsibility for acts or omissions of NGEs as if they had been carried out directly by the State. This highlights the need for States to act robustly when authorizing and continually supervising the activities of NGEs to ensure compliance with the provisions of the OST and international law. Here, Musk’s actions do not constitute a violation of the United States’ neutrality obligations, with different facts a commercial actor could bring a non-state party into a conflict.

***

Squadron Leader Tara Brown is a military professor at the Stockton Center for International Law at the U.S. Naval War College, where she co-teaches a course on Air, Space and Cyber Law.

Photo credit: SpaceX