22 Jul Information Operations in International Humanitarian and Criminal Law: Reflections on the Oxford Statement
[Eian Katz is a Legal and Policy Analyst at Canmore Company. He previously served as Counsel at Public International Law and Policy Group.]
Earlier this month, the Oxford Institute for Ethics, Law and Armed Conflict (ELAC) released a statement on “The Regulation of Information Operations and Activities,” marking an important step in the global effort to reckon with the implications for international law of disruptive forms of online speech. The Oxford Statement takes a very broad understanding of “information operation[s] and activities”—“any coordinated or individual deployment of digital resources for cognitive purposes to change or reinforce attitudes or behaviors of the targeted audience.” This definition encompasses both benign and noxious behavior. As for the latter, the two primary forms of problematic speech animating the discussion are disinformation and hate speech (which is mostly referenced in international law in connection to incitement).
The Oxford Statement sets forth ten principles delineating the relation of information operations and activities to various aspects of international law, including sovereignty and nonintervention (Article 2), human rights (Articles 3–7), jus in bello (Article 8), and international crimes (Article 9). These principles are stated at a fairly high level of generality, inviting more detailed analysis. Several areas have indeed already been the subject of some critical scrutiny. The Special Rapporteur on the promotion and protection of the right to freedom of opinion and expression, for example, has recently published thorough reports on both disinformation and hate speech from a human rights perspective. Other qualified commentators have considered whether information operations could violate the principles of sovereignty and nonintervention, or even amount to a use of force.
With these areas of law having already received some attention, this post focusers on the undertheorized treatment of information operations in international humanitarian law (IHL) and international criminal law (ICL), the subjects of Articles 8 and 9 of the Oxford Statement. Drawing from research on wartime disinformation presented in greater detail here, it highlights numerous dilemmas and ambiguities with respect to each in order to motivate further study and discourse.
Information Operations and IHL
The conduct of information operations or activities in armed conflict is subject to the applicable rules of international humanitarian law (IHL). These rules include, but are not limited to, the duty to respect and ensure respect for international humanitarian law, which entails a prohibition against encouraging violations of IHL; the duties to respect and to protect specific actors or objects, including medical personnel and facilities and humanitarian personnel and consignments; and other rules on the protection of persons who do not or no longer participate in hostilities, such as civilians and prisoners of war.—Oxford Statement, Article 8
Information operations pose unique threats to civilians during armed conflict. They may expose them to retaliatory violence, as has been documented in Myanmar, Rwanda, and Syria. They may cause mental anguish, evoking extreme fear, grief, paranoia, or other painful emotions. And disinformation concerning vital human needs, such as medical care or humanitarian assistance, may further jeopardize civilian welfare.
The severity of these harms conflicts with the traditional characterization in IHL of misinformation as a permissible “ruse of war”—”acts which are intended to mislead an adversary or induce him to act recklessly but which infringe no rule of international law applicable in armed conflict and which are not perfidious.” Information operations today contrast conspicuously with this category in that they are not necessarily directed at a military opponent, may in fact be in breach of other rules of international law, and differ both in degree and in kind from the forms of misinformation imagined by the drafters of the Additional Protocols.
Accepting that information operations should not be allowable per se as ruses of war, an immediate problem that emerges in applying IHL to them is that the harms they cause are often inflicted indirectly. That is, with the exception of mental harms, disinformation and hate speech only result in cognizable injuries to civilians when they or another party choose to act upon it. The involvement of intervening actors who are not under the direction or control of the principal makes it unlikely that these acts would be attributed to the publisher, even if the gravity of the resultant harms might otherwise trigger IHL protections for civilians against attacks and other forms of violence.
Common Article 1 offers a potential solution to this problem. In it, the parties to the Geneva Conventions pledge not only to respect the laws of war themselves, but to “ensure respect” for them by other parties. The International Committee of the Red Cross (ICRC) clarifies that this includes “private persons whose conduct is not attributable to the State,” thus establishing a positive duty incumbent upon combatants to account for all foreseeable violations. However, the scope of this obligation is limited to actors over whom a party “exercises authority.” This may be a significant restriction with respect to online information operations, which can be conducted from afar. It is also worth noting that the positive Common Article 1 obligation to enforce compliance with IHL is coupled with a negative duty not to encourage violations through direct advocacy.
IHL may offer protection to civilians from the direct effects of information operations in several different ways. First, the Tallinn Manual explains that “severe mental suffering” may be counted among the injuries that would qualify an act as an attack. Mental harms are frequently neglected in international law and only loosely defined—while they “need not
permanent and irremediable,” they must go “beyond temporary unhappiness, embarrassment, or humiliation… result[ing] in a grave and long-term disadvantage to a person’s ability to lead a normal and constructive life.” Notably, IHL also prohibits acts intended to “spread terror” among the civilian population, but only when that is their “primary purpose.” This may not be true of information operations, which are instead commonly aimed at sowing confusion, deflating morale, or punishing opponents. Indeed, some information operations may themselves be forms of collective punishment. The ICRC holds that this prohibition covers “penalties of any kind” and is to be given the widest possible interpretation. But internationaltribunals have read in a requirement of intent, such that only disinformation or hate speech undertaken for the purpose of retribution would be deemed collective punishments. Finally, IHL features a number of protections that apply during “militaryoperations,” a term that is definedinconsistently in the ICRC commentaries but which may encompass information operations.
Along with its effects on protected persons, disinformation may undermine the purpose of civilian objects, such as hospitals or humanitarian facilities, by leading to the imposition of operational restrictions upon them or by dissuading civilians from making use of them. In Ukraine, for instance, Russian disinformation about COVID-19 may have led civilians to forego medical care. In some cases, “impairing the functionality of an object” may be a form of attack, though perhaps only if it “requires the replacement of physical components” to restore operability. Alternatively, IHL forbids combatants from acts that “render useless objects indispensable to the survival of the civilian population,” namely, food and water installations. Information operations baselessly disparaging these objects might run afoul of this provision.
A final challenge to the application of IHL to information operations is the frequent engagement in such activities of not only combatants, but civilians and civilian objects as well. Attenuation in the causation of harm (save for mental harm) once again makes it unlikely that this would be considered direct participation in hostilities. Even if it were, the result would only be the temporary revocation of protections during the period of time in which non-military actors take part in information operations, which may be vanishingly brief in the context of online activity. Nor is there a clear method for the determination of the duration of direct participation amid repeated occurrences, leaving a significant gap in the regulatory power of IHL with respect to civilian-led information operations.
Information Operations and ICL
Conducting information operations or activities will amount to international crimes, such as genocide, including direct and public incitement thereto, war crimes and crimes against humanity, where the elements of those crimes are fulfilled.—Oxford Statement, Article 9
Information operations raise the specter of several international crimes for the individuals responsible for them. The Oxford Statement mentions one of these specifically: direct and public incitement to genocide. The need for such incitement to be “direct” separates pure disinformation from hate speech and advocacy of violence; that it also be “public” is axiomatic of information operations. The cases of Myanmar and Rwanda grimly illustrate how information operations might form the basis for this offense.
It is somewhat less likely that information operations would be charged as war crimes, many of which are premised on the occurrence of an “attack.” Due to the issues relating to attribution discussed above, disinformation and hate speech may not qualify as such except as pertains to mental suffering. The war crimes of “wilfully causing great suffering, or serious injury to body or health” and “cruel treatment” (i.e. inflicting “severe physical or mental pain or suffering”) capture this form of harm. One potentially applicable war crime that does not require an attack is “starvation as a method of warfare by depriving [civilians] of objects indispensable to their survival.” As noted above in reference to objects which are rendered useless or non-functional, it is possible that this would be accomplished through a disinformation campaign.
An “attack” is also a prerequisite for crimes against humanity, but not in the same sense as in IHL and war crimes. Instead, it may denote any “any mistreatment of the civilian population” that “caus[es] physical or mental injury,” including by “exerting pressure on the population to act in a particular manner.” Information operations may match this description, and would also probably satisfy the other chapeaux conditions of being “widespread” or “systematic” and “pursuant to or in furtherance of a State or organizational policy.” If so, they might be prosecuted as the crime against humanity of “other inhumane acts,” either for directly causing mental suffering or for indirectly leading to the denial of or abstention from food, water, or medical care.
Lastly, even if not committing substantive offenses themselves, the architects of information operations may accrue accessorial liability for the criminal acts of others through three modalities: solicitation and inducement, aiding and abetting, and contributions to group criminality. First, courts have in the past viewed some forms of hate speech, like derogatory language toward a minority group and violent rhetoric, as a form of solicitation or inducement. Second, and perhaps less realistically, information operations would aid and abet crimes if they were designed to and in fact did facilitate offenses committed by actors who had already been plotting criminal activity. Third, secondary actors may be accountable for acts that significantly contribute to group criminality when they were previously aware of the group’s criminal intentions. Charges against media figures propagating disinformation have been brought before the ICC under this theory in at least two cases.
In the era of hybrid warfare, the ascendant threat of information operations and activities poses a pressing and vexing challenge to international law, casting into vivid relief a host of unresolved questions. The Oxford Statement provides helpful direction to the legal community in grappling with this phenomenon, but only sparse detail. And while scholars have been laboring to fill observable gaps in the application of the law to information operations, many remain. In IHL, difficulties persist with respect to the attribution of physical harms occasioned by information operations, the territorially-bounded scope of Common Article 1, the threshold and evidentiary standards for mental injury, the meaning of “military operations,” and the limitations of “direct participation in hostilities” in addressing the involvement of civilians. Depending on the facts, ICL may have a prominent role to play, especially as concerns incitement to genocide, war crimes relating to mental suffering or starvation, the crime against humanity of “other inhumane acts,” and the various modes of accessorial liability. Mapping the contours of these rules as they relate to information operations will be a vital project for international law in order to restrain its malign effects.