Symposium on Forensic and Counter-Forensic Approaches to International Law: Remote Sensing of Mass Atrocities From Space – How Satellite Capabilities May Qualify Knowledge in Complicity and Trigger Duties to Ensure Respect

[Pedro R. Borges de Carvalho is a PhD candidate at KU Leuven, Institute for International Law and a research fellow at ATHENE – German National Research Center for Applied Cybersecurity]

Mass atrocities often leave scars on the Earth’s surface that forensic methodologies can decrypt given the right technological capabilities. High-resolution satellite imagery analysis is one such method, and it is fundamental for both forensics andcounter-forensics. Its first prominent use in the documentation of an atrocity came in 1995, when the US published its military satellite imagery of the Srebrenica genocide. Before-and-after shots of the devastated Bosnian town evidenced mass executions of its Muslim population after an attack by the Bosnian Serb Army. The imagery was circulated too late, one of many intelligence blunders that spurred years of acrimonious debate over US and NATO responsibility for their failure to prevent that genocide 30 years ago.

A key issue remains as real today as it was then: the selective omniscience of States with their satellite intelligence. Imagery is collected in droves by capable States and commercial satellite operators under their jurisdiction, but the decision of what imagery to actually analyse and interpret through time (“diachronic omniscience”), and which to ignore (“blind omniscience”), is one resting on States’ own political interests.

While NGOs have leveraged the advent of commercial satellite imagery for pursuing international criminal justice, civil society’s potential for autonomous Mass Atrocity Remote Sensing (MARS) remains severely limited by economic and operational constraints. The costs of acquiring near real time imagery from a commercial provider is prohibitive for most civil society organisations. Even where capital may be available, the granularity of the imagery commercial satellites are allowed to provide is still insufficient, since most States legally restrict Very High Resolution (VHR) imagery capabilities to government agencies, as the US does specifically in relation to imagery of Israel and its occupied territories: under US law, non-federal entities are banned from acquiring imagery ‘more detailed or precise than satellite imagery of Israel that is available from commercial sources’. And even where the collection of adequate imagery by humanitarians is possible, the specific skills required for making sense of it (imagery analysis) is in short supply, and mostly found within State intelligence services and other specialised organs of State. Factually, and with due merit to some notable exceptions, this is still the domain of States.

Yet, even among States, satellite imagery is a limited resource. Distribution of technological capabilities is sharply uneven among States, and the determination of what territory to watch and when is subject to strategic priorities and bottlenecked by image analysis limitations. At the extreme, the orbital slots and frequency spectrum themselves “are limited natural resources and must be used rationally, efficiently and economically”, pursuant to Article 44(2) of the ITU Constitution.

Hence, monitoring Earth from space is as critical as it is difficult, and ranking interests becomes unavoidable. From here, this post sets out to forward certain interests it deems imperative pursuant to international law, and considers what consequences flow from possessing a higher level of technological prowess in geospatial intelligence (GEOINT) in connection with these interests – namely, the prevention and punishment of serious breaches of international humanitarian law (IHL) and other peremptory norms.

If possessing “eyes in the sky” confers GEOINT-capable States the (rare) potential to detect and document mass atrocities everywhere and at all times, does this entail legal consequences for such States? That is, can this capability warrant an inference of knowledge of wrongful circumstances where mental elements are concerned in the law of State responsibility – notably in complicity? And may GEOINT powers, such as the United States (a State concentrating over 5000 of all roughly 7,500 operating satellites in the world), become obligated to document evidence of certain atrocities under the duty to ensure respect for IHL? If so, at what stage in time?

Feigning Ignorance: Inferred Knowledge and Wilful Blindness in Complicity

Under the customary international law of State responsibility, the mode of attribution of aid or assistance (or “complicity”) assigns responsibility to an aiding State for its assistance in the wrongful act of another State. This customary rule of attribution is codified in Article 16 of the Articles on Responsibility of States for Internationally Wrongful Acts (ARSIWA): “A State which aids or assists another State in the commission of an internationally wrongful act by the latter is internationally responsible for doing so if: (a) that State does so with knowledge of the circumstances of the internationally wrongful act; and (b) the act would be internationally wrongful if committed by that State.” Complicity is a difficult mode of attribution because it encompasses an elusive subjective element (knowledge), and because what counts as assistance is not always clear. This short post purports only to address the debate to a limited extent, focusing on the matter of high-resolution satellite capabilities in relation to complicity’s mental (or subjective) and material elements.

In assessing the mental element to complicity, i.e. whether an assisting State had knowledge of the wrongful act pursued by the principal State, there are three main possible standards. The first is the typical evidentiary inference drawn by the International Court of Justice (ICJ) in Corfu Channel – that the State must have known about the circumstances. The second is the wilful blindness scenario (true wilful blindness, as per Milanovic), where the assisting State deliberately avoids acquiring information that would produce practical certainty of the circumstances of the wrongful act. To prove wilful blindness, one must prove the ulterior purpose of avoiding truth for avoiding responsibility on the part of the complicit State. But there is also an in-between theory: proof of knowledge through establishing a failure of due diligence in discovering reasonably foreseeable circumstances, i.e. constructive knowledge (the State should have known). There seems to be a blurred continuum between constructive knowledge and wilful blindness.

First, it is important to note in what temporal and circumstantial conditions knowledge in complicity would really be an issue. The types of wrongs with which we are concerned here – serious breaches of jus cogens – would dispense with the need for inferences of knowledge, as knowledge is automatically presumed. As per the International Law Commission’s (ILC) Commentary to Article 41(2) ARSIWA, “it is hardly conceivable that a State would not have notice of the commission of a serious breach by another State.” This sort of presumption of State knowledge seems to mirror the case law of early international claims commissions on massive events of general repercussion that could not have gone unnoticed by organs of State.

What interests us here is the grey temporal line in the development of an atrocity campaign. By what moment does a situation first amount to a jus cogens breach, and hence must necessarily be deemed to have been known by all States in their dealings with the perpetrator State? 

Helmut Philipp Aust and Prisca Feihle suggest deploying due diligence in interpreting knowledge under Article 16 ARSIWA as well as the progression of complicity under Article 14 ARSIWA as the principal breach continues through time. They contend that it “is the time between the beginning of the serious breach and the moment when the consolidation of the outcome of this breach is at stake that the notion of due diligence might have a fruitful role to play” (p. 57) in the interpretation of the subjective element of complicity. Presumably, this could mean interpreting knowledge here in light of a constructive knowledge standard, scrutinising both the feasibility (a State’s technological capabilities to evaluate its cooperation with the principal wrongdoing State, including its GEOINT capabilities) and reasonableness of measures that could be required from that State to foresee the wrongful circumstances at hand.

I would add that the period preceding the configuration of a situation as a breach of jus cogens, where presumption of knowledge of Article 41(2) ARSIWA could not yet be properly raised, would particularly benefit from this approach to knowledge. At this stage, pursuant to a due diligence standard, a State may need to,for example, direct its satellite intelligence analysts to maintain a “detection posture”, working to assess whether there is a risk of an atrocity being committed by its partner State. The period referred to by Aust and Feihle is what happens afterwards, once the jus cogens violations materialise. When this becomes verifiable by the assisting State, besides needing to reevaluate its cooperation to stave off attribution of responsibility under Art. 16 ARSIWA, it must shift its GEOINT stance from detection to a “documentation posture”, gathering evidence of the atrocities for future redress.

It is at this point that the issue transitions from knowledge of another State’s wrongful act for the purposes of determining complicity to the assisting State’s own obligations of conduct. In terms of complicity’s material element, it is yet debated whether an omission could qualify as aid and assistance. The ILC itself never expressly said that complicity had to come in the form of a positive conduct, and Aust has argued that scenarios of complicity by omission are plausible and possible. The more common occurrence, however, is for an omission to configure a breach of a self-standing obligation of due diligence, such as the customary duty to ensure respect for IHL, crystallised in Common Article 1 (CA1) of the 1949 Geneva Conventions.

GEOINT Evidence Gathering as a Duty to Ensure Respect or Duty of Non-Assistance

Under customary IHL, States have a duty to “exert their influence, to the degree possible, to stop violations of international humanitarian law.” To the International Committee of the Red Cross and other advocates of the “expansionist camp”, this duty includes an important positive dimension, requiring concrete action by other States toward violating States, which may include, I submit, the collection and analysis of satellite imagery for the prosecution of war crimes.

Defenders of the restrictive camp would rebut that the duty mostly comprises only a negative dimension, requiring States not to assist others in breaching IHL. This obligation of restraint would, so this camp argues, represent a clearly different thing to a positive duty to exert influence.

However, an omission can also distinctly benefit another State in its violations of IHL, to the extent that the framing as positive or negative conduct could become a reproachable exercise of mental gymnastics. Where a State has satellite imagery that constitutes evidence of serious breaches by another State, shelving it means assisting the violating State through its omission. The CA1 duty entails an obligation not to “take action that would assist in” violations of IHL. Withholding evidence that few others could produce (such as VHR satellite imagery) could well be construed as such a prohibited conduct – an assistance by omission.

Though this post has focused on the CA1 duty due to its better-developed doctrinal tradition, the same rationale seems pertinent for interpreting the duty of non-assistance in serious breaches of jus cogens generally (Article 41 ARSIWA), including genocide, crimes against humanity, aggression and the right to self-determination of peoples.

Concluding Remarks

This post has attempted to shed light on the continuum between complicity and non-assistance by touching on the various axes of time, knowledge and power underpinning these legal constructs. It has attempted to do so from as pragmatic as possible a standpoint. It has not claimed that States ought to become omniscient of all core crimes committed on Earth at any time, as one cannot expect the performance of the impossible. Rather, it has explored under what circumstances States endowed with certain GEOINT technological prowess will be deemed aware of core crimes which they could not reasonably have remained unaware of, and when they would be required to assume a detection or documentation posture of such events from Space. This could mean watching a territory at known risk of a genocidal campaign (the reason why satellite imagery of mass graves in Srebrenica was captured to begin with), documenting internment camps and widespread destruction of temples, or yet detecting the amassment of troops at another State’s borders as an early sign of the crime of aggression.

That other States refuse to avert their remote sensing apparatuses is not only necessary for punishing serious breaches – it is also essential for fostering international accountability in the now and dissuading potential perpetrator States from new or continued core international crimes. In this sense, monitoring and documenting developing mass atrocity campaigns is also a preventive conduct with crucial implications, both before and after the catastrophe.