Introduction
According to the World Risk Poll, “fake news” topped the list of concerns for internet users in 2020.Footnote 1 This finding should come as no surprise to those who have followed with growing consternation the proliferation of false or misleading information dirtying the arenas of politics,Footnote 2 health,Footnote 3 financeFootnote 4 and other sectors. Less noticed amid the handwringing have been the unique effects of fake news on civilians in armed conflict settings. Whereas belligerents have long recognized the military value of disinformation for deceiving their enemies, in the age of hybrid warfare it is increasingly produced for a civilian audience. Positioning civilians as the subject or object of disinformation or propaganda may fan the flames of violence and may further lead to three distinct and cognizable forms of civilian harm.
First, fabrications that vilify disfavoured individuals or groups may serve to encourage and legitimate foreseeable acts of violence perpetrated against them by third parties. Perhaps the most infamous such episode occurred in the prelude to the Rwandan genocide, when radio stations owned by the State or by high-ranking State officials inflamed popular sentiment against the Tutsis through fake news reports.Footnote 5 In other conflicts, organized smear campaigns have likewise endangered minority populations or humanitarian workers.Footnote 6
A second unsettling application of disinformation that conflict parties have recently embraced is in distorting information vital to securing human needs. Through widespread fearmongering and deliberate obfuscation, hybrid actors aim to sow dissent, undermine the social order, aggravate crises and discredit enemy institutions, including civilian ones. Throughout the COVID-19 pandemic, for instance, various conflict actors have spread disinformation about the virus, underreported case counts or boosted conspiracy theories for political and economic gain.Footnote 7 Disinformation in armed conflict may likewise be directed against humanitarian organizations providing relief to beleaguered civilians, jeopardizing their operations and sullying their reputations.Footnote 8 These rampant misrepresentations can place civilians in harm's way by disrupting access to services or by leading them to act in ways counter to their own interests.
Finally, disinformation may directly harm the mental health of civilians by arousing extreme fear, grief or other painful emotions or unsound mental states. Disinformation geared toward civilian audiences might lead them to develop paranoia or conspiratorial thinking, doubt in their continued ability to satisfy their human needs, a belief that friends or relatives have been or will be harmed, or a reasonable apprehension of death or bodily injury. These harms are less amenable to documentation and proof, but may be no less traumatic, damaging and enduring.
Often, disinformation produced during armed conflict differs little in content from that produced in other contexts. Even so, there are salient practical and legal distinctions. Practically, conflict-affected persons may be especially vulnerable to the ill effects of disinformation due to their desperate living conditions, elevated exposure to violence and decreased access to trustworthy sources of information. Compounding the potential for harm is the fact that conflict environments have proven especially fertile incubators for disinformation, as evidenced by its recent flourishing in Israel and Palestine, Libya, South Sudan, Syria, Ukraine and Yemen.Footnote 9 Military and civilian actors alike have made gainful use of disinformation in order to control the narrative regarding the conflict. The deep civil fragmentation characteristic of conflict settings amplifies the effect of such disinformation by creating echo chambers that heighten confirmation bias and accelerate the uncritical sharing of specious reports.Footnote 10 Moreover, the collapse of State institutions amid conflict weakens content regulation networks and limits the availability of judicial remedies for confronting libel,Footnote 11 leaving few restraints on the spread of disinformation.
Legally, disinformation during armed conflict is unique in being governed by international humanitarian law (IHL), which has long characterized it as an allowable “ruse of war” (provided that it complies with other applicable rules).Footnote 12 But, it shall be argued, this permissive attitude is the product of a bygone era in which deceptive practices could be safely supposed to yield tactical or political gains without significantly impacting the condition of non-combatants. The contemporary trend of disinformation being deployed in ways that threaten civilian populations challenges that underlying assumption and begs for a re-evaluation of its treatment in IHL.
The primary enterprise of this article, therefore, is to pioneer a new legal approach to disinformation in armed conflict by mining the corpus of international law for rules that might constrain it. The first part of the article begins with a review of the traditional IHL stance towards disinformation and contends that this stance is no longer in keeping with disinformation's modern manifestations. After discussing two preliminary issues in the second part, the remainder of the article proceeds to search for other applicable protections with respect to the three forms of harm described above: exposure to foreseeable acts of violence, distortion of information concerning vital human needs, and mental suffering. In most cases, the applicability of protections under IHL, international human rights law (IHRL) or international criminal law is speculative and incomplete at best, calling for greater clarification in the law and focused attention on addressing this problem.
The scope of this article is confined by two principal limits. First, it is concerned only with the publication of false information, not advocacy of violence. It therefore does not delve deeply into the law of incitement, which is the subject of substantial existing case law and scholarship. Incitement is only examined herein insofar as it might arguably describe incendiary falsehoods published with the intent to provoke a violent reaction, even absent an explicit call for violence. Second, this article only deals with disinformation from a jus in bello perspective. Other qualified commentators have opined on disinformation as a use of force or as a violation of sovereignty,Footnote 13 and have examined its ordinary treatment in IHRL.Footnote 14
A word on terminology: “disinformation” and “misinformation” both refer to false information but are separated by an element of intentionality in the former: disinformation is “deliberately misleading or biased”.Footnote 15 “Fake news” is often a form of disinformation or misinformation but is poorly defined and therefore used sparingly here. Other related concepts, such as “propaganda”,Footnote 16 “psychological operations”,Footnote 17 “influence operations”,Footnote 18 “information operations”,Footnote 19 “information warfare”Footnote 20 and “cognitive warfare”,Footnote 21 may also feature some combination of disinformation, misinformation, misleading information and accurate information. For the sake of clarity, this article avoids them.
Disinformation in armed conflict
The deployment of disinformation as a military tactic is hardly a modern innovation. Writing in the fifth century BC, Sun Tzu maintained that “all warfare is based on deception”. Classical military history is replete with storied battlefield artifices, conjuring the legendary tales of the Trojan Horse, Alexander the Great's surprise crossing of the Hydaspes River, and Hannibal's ambush at Lake Trasimene. In the modern period, subterfuge and deceit remain critical components of military strategy, facilitated by increasingly sophisticated communications technologies.
IHL takes a remarkably lenient approach to such stratagems, with one exception. Combatants may not resort to perfidy, defined in Article 37(1) of Additional Protocol I to the Geneva Conventions (AP I) as “inviting the confidence of an adversary to lead him to believe that he is entitled to, or is obliged to accord, protection under the rules of international law applicable in armed conflict, with intent to betray that confidence”.Footnote 22 Belligerents thus may not, for example, disingenuously accept a truce or announce their capitulation, feign injury or illness, or claim civilian or other protected status.Footnote 23
In the next paragraph of Article 37, however, AP I explicitly permits “ruses 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”.Footnote 24 While this particular provision only pertains to international armed conflicts (IACs), ruses are likewise licensed in non-international armed conflicts (NIACs) under customary international law.Footnote 25
“Misinformation” is expressly classified in AP I as a type of permissible ruse in armed conflict.Footnote 26 Several other authoritative sources confirm this posture throughout IHL. The International Committee of the Red Cross (ICRC) Commentary on AP I categorizes “circulating misleading messages” as a form of ruse,Footnote 27 and numerous national military manuals likewise endorse the use of “misinformation”, “disinformation”, “false information”, “psychological operations” or “bogus dispatches and newspapers”.Footnote 28 The Tallinn Manual on the International Law Applicable to Cyber Warfare (Tallinn Manual), the leading treatise on the application of international law to cyberspace, also enumerates examples of acceptable ruses that include “psychological warfare activities” such as “dropping leaflets or making propaganda broadcasts”.Footnote 29 In customary law, as conveyed by the ICRC, ruses are defined even more expansively than in AP I – as “acts intended to confuse the enemy”Footnote 30 – such that there can be little doubt that misinformation would be included. Following this logic, many scholars have presumed that IHL constrains disinformation in armed conflict only via the rule against perfidy and other background principles such as necessity, proportionality and distinction.Footnote 31
But modern applications of misinformation – or, more accurately, disinformation – call into question its reflexive characterization in IHL as a ruse for several reasons. First, while ruses are presented in source materials as being intended to mislead an “enemy” or “adversary”, disinformation campaigns during armed conflict today are instead often oriented primarily towards the civilian population.Footnote 32 While this does not automatically render them illegal,Footnote 33 it highlights a glaring descriptive deficiency in the traditional IHL conception.
More problematic still for the classification of disinformation as a permissible ruse under IHL is the requirement that such acts “infringe no rule of international law applicable in armed conflict”.Footnote 34 This proviso demands a fuller analysis of the relevant aspects of IHL, IHRL and international criminal law governing the spread of disinformation during armed conflict, an inquiry that this article will take up later on. As shall be shown, certain forms of disinformation propagated during armed conflict may in fact be in breach of international law.
Finally, the inaptness of the definition of disinformation points to larger issues related to the ability of IHL to adapt to technological developments. Modern strains of disinformation differ both in degree and in kind from their progenitors. Mass media facilitates the generation and circulation of disinformation at unprecedented levels, such that sorting fact from fiction becomes exceedingly difficult.Footnote 35 This phenomenon is sometimes referred to as an “infodemic”.Footnote 36 Relatedly, technological advancements have made fabrications appear increasingly realistic.Footnote 37 Thus, whereas the ICRC once supposed a ruse to be “very often the only course open to a weak combatant”,Footnote 38 disinformation today is often employed by highly sophisticated combatants and non-combatants using complex methods. For these reasons, it no longer makes sense to label disinformation as a “ruse of war” without fuller consideration of the circumstances.
Preliminary considerations
Accepting the contention that disinformation during armed conflict should not be afforded a blanket permit under international law as a ruse of war, it must be considered how else it is to be governed. The remainder of this article explores this topic. It finds that there is no single rule, but rather a patchwork set of protections cobbled together from different branches of international law that differ substantively with respect to the aforementioned three categories of harm: exposure to foreseeable acts of violence, distortion of information concerning vital human needs, and mental suffering. But the straitjacketed application of these protections leaves considerable legal blind spots through which disinformation may continue to exact a civilian toll.
Before embarking upon this legal analysis, two threshold issues must be addressed. The first is whether the protections of IHRL are applicable in this context at all, and if so, with what limitations. The second is whether persons and objects ordinarily protected by IHL surrender their privileged status by participating in disinformation operations.
Applicability of IHRL
Several technical considerations complicate the continued applicability of human rights law during armed conflict. The first is its relationship to IHL. During armed conflict, IHRL and IHL apply concurrently.Footnote 39 In cases of overlap or conflict between them, however, IHL generally prevails as lex specialis.Footnote 40 Alternatively, when one corpus does not address a particular issue, the other governs by default. It is asserted here that the provisions of IHL accepting ruses of war do not apply to disinformation in its modern form.Footnote 41 The parts of the article that follow further demonstrate that, on the whole, there are few other IHL rules regulating such disinformation. This relative silence should admit of greater reliance on IHRL in this space.
The involvement of non-State actors or of States acting outside their own territories in producing disinformation during armed conflict introduces additional wrinkles to the analysis. IHRL is primarily addressed to sovereigns, and non-State actors are usually not counted among the subjects of human rights treaties.Footnote 42 Nonetheless, non-State actors are bound at minimum to abide by jus cogens norms,Footnote 43 such as the right to life.Footnote 44 Moreover, the United Nations (UN) considers non-State actors to be subject to IHRL when they exercise governmental functions or exert control over a territory or population.Footnote 45 But there is no clear legal source for these imputed obligations, leaving the applicability of IHRL to non-State actors an unresolved question.Footnote 46 While non-State groups that produce disinformation would therefore be required to respect the right to life everywhere as jus cogens, the effect of other human rights standards within their areas of control is a matter of debate.
States may also produce disinformation for consumption by civilians beyond their territorial limits. Whether IHRL limits this activity depends upon the theory of extraterritorial jurisdiction applied. International courts have mostly hewn to the “effective control” or “personal jurisdiction” models,Footnote 47 according to which it is unlikely that human rights law would apply in these circumstances. Some scholars have instead advocated for frameworks based upon the distinction between positive and negative rightsFootnote 48 or a functional analysis.Footnote 49 These more flexible approaches would be more accommodating of extraterritorial jurisdiction in human rights law as pertains to wartime disinformation operations, but they are merely de lege ferenda at present.
Disinformation as direct participation in hostilities
Combatants are not the only actors who have weaponized information in armed conflict. Public officials, bureaucrats and civilian branches of government may be involved in the production and dissemination of disinformation. Private citizens and other non-State actors, whether acting in the service of the conflict actors or in their individual capacities, also have a prominent role, and in some cases have been shown to be the primary drivers of disinformation in conflict.Footnote 50 In several contemporary armed conflicts, the combatants have deputized “electronic armies”, “troll farms” or “web brigades” to spew disinformation.Footnote 51
Civilians and civilian objects – defined in the negative as, respectively, persons who are not members of the armed forcesFootnote 52 and objects which are not military objectivesFootnote 53 – are afforded a protected status under IHL, but it is not immutable. Civilian protections may be forfeited by direct participation in hostilities, which consists of three elements: (1) a minimum threshold of harm, including “death, injury, or destruction”; (2) “a direct causal link between the act and the harm”; and (3) the act must be undertaken in support of a party to the conflict (the so-called belligerent nexus).Footnote 54
Disinformation harming civilian life or civilian objects during armed conflict may satisfy the first and third elements, but it is likely to fail on the causal element, save for mental injuries inflicted directly. According to the ICRC, direct causation means involvement in the actual conduct of hostilities, not merely contributing to the war effort or to war-sustaining activities.Footnote 55 Disrupting the supply of electricity, water or food, as disinformation concerning vital human needs might, does not qualify as direct participation in hostilities.Footnote 56 Nor does political propaganda,Footnote 57 which has analytical similarities to disinformation. On the other hand, publishers may become legitimate military objectives if they engage in incitement.Footnote 58 Depending on the context, disinformation may be more akin to propaganda, incitement, or neither.
Even if publishing disinformation were in some cases treated as direct participation in hostilities, it is difficult to imagine combat action being directed against individual social media users. Civilian objects engaged in disinformation, such as broadcasting stations or government bureaus, would make for more realistic targets. However, the protected status of civilian objects would only be revoked for the period of time during which they were producing disinformation, which may be very brief in each instance.Footnote 59 Nor is there a clear standard in IHL defining the duration of direct participation amid repeated occurrences,Footnote 60 such as by frequent publishers of disinformation. For these reasons, it would be exceedingly difficult to legally justify military operations against civilians or civilian objects directly participating in disinformation.
Exposure to foreseeable acts of violence
This article now turns to its core project: evaluating the restrictions placed by international law upon disinformation in armed conflict with respect to the various types of civilian harm that it may cause. This part of the article directs its scrutiny toward the first type: disinformation about an individual or group that is liable to inspire others to react violently against them. Few legal rules restrain this behaviour. Despite causing harm of a magnitude accordant to an “attack”, the inherent attenuation in agency makes it unlikely that this form of disinformation would be designated as such under IHL. Instead, disinformation that leads to foreseeable violations of the Geneva Conventions would be prohibited under common Article 1, but only with respect to perpetrators over whom the conflict actor “exercises authority”. Nor are disinformation operations likely to be treated as incitement unless they are accompanied by advocacy of violence. That said, it is possible that the individuals responsible for them could be charged as criminal accessories.
Violence and the problem of indirect harm
Among the most fundamental rules in IHL is that civilians “shall not be the object of attack”.Footnote 61 Were disinformation deemed an “attack”, therefore, civilians and civilian objects would be legally shielded from foreseeable and non-incidental injuries, deaths, destruction or damage caused by it.Footnote 62 The conflict parties would likewise be barred from employing disinformation indiscriminately, without a specific military objective in mind.Footnote 63 Furthermore, civilians in armed conflict are generally protected from murder, torture or other violence to life or well-being.Footnote 64
Attacks are defined in AP I as “acts of violence against the adversary, whether in offence or defence”.Footnote 65 “Violence” does not only mean uses of physical force;Footnote 66 possible attacks are instead commonly evaluated with reference to their expected effects. For persons, these effects must include “injury or death” in order to be considered an attack.Footnote 67 As confirmed by the cases of Myanmar, Rwanda and Syria,Footnote 68 disinformation in armed conflict may very well cause these sorts of harms to occur, but the inability to attribute these harms to the publisher makes it unlikely that disinformation would be treated as an attack or other act of physical violence under IHL.
Disinformation defaming particular individuals or groups only leads to physical injury when others choose to act upon it. To say that the publisher has attacked the civilians in question would therefore be to attribute to it the violent response of third parties. There are certain circumstances in which harms may in fact be ascribed to the principal despite the involvement of intervening actors. The conduct of persons or groups acting under the instruction, direction or control of a State, for instance, legally accrues to the State.Footnote 69 But disinformation by itself does not ordinarily entail direction or compulsionFootnote 70 to commit an offending act; it simply supplies or reinforces the motive.
Another way to look at the relationship between disinformation and civilian harm would be through the lens of causation. As described above, in the context of direct participation in hostilities, the ICRC has construed direct causation to require that “the harm in question must be brought about in one causal step”.Footnote 71 Merely providing the motive to effect harm, as disinformation might, is not enough.Footnote 72 Another view that has been put forward is the notion of an “indirect attack” committed by combatants who wilfully expose civilians to danger from the otherwise lawful action of their adversaries, such as by using them as human shields.Footnote 73 Disinformation may analogously place civilians at risk of violence exacted by third parties, though not the sort that would ever be legally authorized. It thus seems unlikely, for reasons related to attribution and causation, that disinformation would be labelled an unlawful attack or other physically violent act.
Common Article 1
In common Article 1, the parties pledge to “ensure respect” for the Geneva Conventions. The ICRC Commentary interprets this duty as applying to “the whole population over which a High Contracting Party exercises authority”, including “private persons whose conduct is not attributable to the State”.Footnote 74 Combatants must, therefore, “take measures to ensure respect for the Conventions by private persons”.Footnote 75
The obligation to “ensure respect” bears both positive and negative aspects. On the positive side of the equation, conflict parties must affirmatively act to prevent foreseeable violations.Footnote 76 This might plausibly be read to enjoin the publication of disinformation when it is foreseeable that it would lead to violations of IHL, regardless of the number of links in the causal chain. But common Article 1 is confined in scope to persons over which a party “exercises authority”, likely requiring some form of control or jurisdiction. Thus, a conflict actor may not be responsible under this rule for disinformation produced within its domain that causes third parties beyond its area of control to violate IHL.
In the negative, common Article 1 is also held to provide that the parties are to refrain from “encourag[ing]” violations of the Conventions.Footnote 77 The ICRC derives this duty from the International Court of Justice's (ICJ) judgment in the Nicaragua case, wherein the United States distributed manuals openly calling on the rebels to commit acts of violence.Footnote 78 The facts of the Nicaragua case therefore suggest that the encouragement which the ICRC had in mind exhibits actual advocacy of wrongdoing, not mere disinformation.
Incitement
While human rights treaties do not directly forbid State signatories from engaging in incitement to violence, they are obligated to enact legislation outlawing it.Footnote 79 Many States have fulfilled this duty, which is closely tied to the right to life,Footnote 80 by passing statutes criminalizing incitement.Footnote 81
Incitement may be distinguished from other expressive acts by the application of a six-factor test.Footnote 82 In the context of disinformation, the two most important factors are intent and likelihood. The element of intent differentiates actual “advocacy” of discrimination, hostility or violence from the “mere distribution or circulation of material”.Footnote 83 Disinformation that takes the form of libellous accusations against individuals or groups likely does not reach this level, no matter how provocative, unless it specifically promotes a violent course of action. Even if it does, the publication would not be considered incitement unless there is also a “reasonable probability” that the allegations therein would directly lead to harm, regardless of whether or not such harm ultimately materializes.Footnote 84 These requirements make it rather unlikely that disinformation which does not itself commend violence, even if implicitly designed to spark outrage, would be considered incitement under IHRL.
In a small subset of cases, incitement may also be a criminal matter. Specifically, the Rome Statute of the International Criminal Court (ICC) criminalizes “direct” and “public” incitement to commit genocide.Footnote 85 Case law from the International Criminal Tribunal for the former Yugoslavia (ICTY) and International Criminal Tribunal for Rwanda (ICTR), whose statutes included similar provisions, reveals that “direct” means an actual invitation, whether explicit or implicit, to commit genocidal acts.Footnote 86 It is therefore again dubious that disinformation alone would ever be prosecuted as incitement to genocide.
Accessorial liability
Even if not violating international law themselves, those propagating disinformation likely to elicit a violent response may accrue vicarious or accessorial liability for the criminal acts of others. International law recognizes several different forms of accessorial liability, which are compiled in Article 25(3) of the Rome Statute. These include solicitation and inducement, aiding and abetting, and contributions to group criminality.
Solicitation and inducement
Liability for a criminal offence may extend to the person who “orders, solicits, or induces” its commission.Footnote 87 These acts each entail “prompting another person to commit a crime” and are referred to collectively as “instigating”.Footnote 88 “Ordering” differs from the latter two modalities in that it implies a relationship of authority between the accessory and the principal,Footnote 89 which likely would not exist in a case of disinformation. Inducement involves an “exertion of influence by the accessory over the physical perpetrator”, whether by rhetoric or action.Footnote 90 Solicitation is a lower standard, whereby the accessory merely “asks for” the commission of the crime.Footnote 91 In any case, the instigation must have a “direct effect” on the commission or attempted commission of the crime, such that the instigator is its “intellectual author”.Footnote 92 The instigator must also have been aware that the resultant conduct would occur in the ordinary course of events.Footnote 93
In case law, solicitation and inducement often involve recruitment or instruction to commit a specific act.Footnote 94 However, more general speech acts have also been treated at times as indicative of instigation, such as derogatory language toward a minority group,Footnote 95 campaign activities employing violent rhetoric,Footnote 96 and “public statements which indicate an intention to hold on to power at any cost, including by use of force against civilians”.Footnote 97 Disinformation that tends to arouse popular animosities towards identifiable targets might be viewed similarly.
Aiding and abetting
A second mode of co-perpetration in international law is “aid[ing], abet[ting] or otherwise assist[ing]” in the commission or attempted commission of a crime.Footnote 98 This is associated with a lower degree of culpability than instigation because the principal is determined to commit the crime before the intervention of the accessory.Footnote 99 “Aid[ing]” and “otherwise assist[ing]” are given the same meaning: providing “practical or material assistance”.Footnote 100 “Abet[ting]”, by contrast, connotes “moral or psychological assistance”, including “encouragement of or even sympathy for the commission of the particular offense”.Footnote 101 While there is no “minimum threshold”, the assistance must have “furthered, advanced or facilitated” the criminal conduct.Footnote 102 Finally, according to the Rome Statute, the perpetrator must have acted with “the purpose of facilitating the commission of [the] crime”,Footnote 103 a stricter mens rea requirement than is found elsewhere in the Statute.
Liability for aiding and abetting has mostly been found for defendants who in some way actively take part in the criminal conduct or bolster the principal's capacity to commit the crime. Culpable behaviour may include providing logistical or communications support, knowingly transferring illicit payments, or participating in planning, recruitment or training.Footnote 104 Moral support may be expressed through physical presence or by failing to object when informed of the crime.Footnote 105 Based on these precedents, a disinformation campaign would aid and abet offences only if it were specifically designed to embolden actors who had already been plotting criminal activity.
Contributions to group criminality
The final form of accessorial liability recognized by the Rome Statute is for acts which “[i]n any other way contribute[ ] to the commission or attempted commission of … a crime by a group of persons acting with common purpose”.Footnote 106 This is a residual mode of secondary liability intended to capture culpable acts other than ordering, soliciting, inducing or aiding and abetting.Footnote 107 It is also directed specifically toward group criminality,Footnote 108 which might include the acts of States or organized armed groups involved in disinformation operations. In order for liability to attach, the contribution must be “significant”,Footnote 109 intentional, and either in furtherance of the criminal purpose or with knowledge of the group's criminal intentions.Footnote 110
Charges against media officials peddling disinformation amid conflict have been brought before the ICC under this theory in at least two cases. In Kenya, charges were confirmed against the head of a radio station accused of “broadcasting false news regarding alleged murder(s) of Kalenjin people in order to inflame the atmosphere”.Footnote 111 In the Democratic Republic of the Congo, the executive secretary of an armed group conducted an international disinformation campaign through radio communications and press releases with the goals of skirting accountability and winning political and military concessions.Footnote 112 This charge was not confirmed for a number of evidentiary reasons.Footnote 113 These cases indicate that individuals responsible for disinformation during armed conflict may be held criminally liable if their dispatches significantly contribute to crimes perpetrated by the belligerents and are intended or known to be likely to do so.
Distortion of information concerning vital human needs
Disinformation concerning health care, food and water supplies, or other essential needs represents a second form of civilian harm that may be brought about by disinformation in armed conflict. From a human rights perspective, these deliberate misrepresentations may violate the rights to health, food, clothing, shelter or water, or the putative right to humanitarian assistance. By discouraging civilians from availing themselves of institutions providing these necessities, such disinformation may also illegally “render [them] useless” and might furthermore be a type of collective punishment. A separate set of IHL protections would apply if the disinformation campaigns were construed as “military operations”. Finally, infodemics that effectively deprive civilians of the information required to make responsible decisions related to their well-being could conceivably constitute a crime against humanity or even the war crime of starvation.
Fundamental guarantees and human rights
Under IHL, civilians in armed conflict are entitled to respect for person and humane treatment,Footnote 114 including medical care “to the fullest extent practicable and with the least possible delay”.Footnote 115 At a minimum, respect for person includes all individual rights, and humane treatment is to be interpreted in its “broadest sense”.Footnote 116 But the content of these “fundamental guarantees” is mostly supplied by other branches of international law, primarily, IHRL.Footnote 117
Disinformation regarding critical services may run afoul of a number of core human rights provisions. During the COVID-19 pandemic, disinformation concerning medical care and medical providers has proliferated. This impinges upon the rights to life and health, both of which impose obligations upon States to prevent and control the spread of diseases.Footnote 118 In order to do so, States must ensure that health information is freely accessible to the public.Footnote 119 Any deliberate “misrepresentation of information vital to health protection or treatment” is a violation of the right to health.Footnote 120
Misinformation concerning humanitarian operations during armed conflict, too, is commonplace.Footnote 121 These misrepresentations may have obvious bearing on the rights to health and life, along with the rights to food, clothing, housing and water.Footnote 122 In addition, some scholars have posited an independent, albeit limited, right to receive humanitarian assistance.Footnote 123 Disinformation regarding access to life-saving goods and services would clearly interfere with this right as well.
Access to information pervades the discourse with respect to all of these rights. In addition to its instrumental role in advancing other rights, access to information is itself substantively protected by the International Covenant on Civil and Political Rights and includes the “freedom to seek, receive, and impart information and ideas of all kinds”.Footnote 124 Information covered by this right includes records retained by public bodies and personal data stored in electronic data files.Footnote 125 But it is not facially apparent that this guarantee restrains the State from publishing inaccurate information or that it would require the State to validate information published by private actors.
One specific “fundamental guarantee” found in the Geneva Conventions and their Additional Protocols is the prohibition of collective punishment.Footnote 126 Collective punishment is considered to consist of “penalties of any kind” and is to be understood in its “widest sense”.Footnote 127 This might reasonably include disinformation designed to harm civilians, though only when undertaken with retributive intent.Footnote 128
Protections for civilian objects
Just like attacks against civilians, attacks against civilian objects are forbidden by IHL.Footnote 129 In this context, an “attack” means acts leading to “damage or destruction to objects”.Footnote 130 “Damage” is not restricted to physical harm, but may also connote “impairing the functionality of an object”.Footnote 131 Following this, it might be argued that certain forms of disinformation have the effect of functionally disabling objects delivering medical or humanitarian assistance in armed conflict by dissuading civilians from making use of them.Footnote 132 While most of the Tallinn Manual drafters considered a loss of functionality to occur only if it “requires the replacement of physical components” to restore operability,Footnote 133 some States have taken a contrary position in their assessment of cyber operations.Footnote 134 France, for example, has stated that a cyber operation constitutes an attack if the targeted object is no longer able to serve its intended purpose, regardless of physical damage.Footnote 135 Disinformation might analogously corrode the utility of civilian objects by altering public perception towards them, albeit through a mechanism less direct than a cyber attack.
In both IACs and NIACs, it is also “prohibited to attack, destroy, remove, or render useless objects indispensable to the survival of the civilian population”.Footnote 136 This provision, which appears alongside the prohibition against starvation as a method of warfare, was conceived as a response to “scorched earth” tactics.Footnote 137 It is therefore primarily applicable to objects vital for sustenance,Footnote 138 namely food and water installations.Footnote 139 While the ICRC Commentary on the Additional Protocols notes that the assortment of verbs (“attack, destroy, remove, or render useless”) is intended to “cover all possibilities”,Footnote 140 examples enumerated in both the Additional Protocols and their Commentary exclusively concern food and water.Footnote 141 In application, too, the scope of “indispensable objects” has been restricted to facilities critical to the food and water supply.Footnote 142 Disinformation that takes the form of discouraging civilians from partaking of such provisions, such as by unjustly questioning their safety, might be thought of as “render[ing] useless” the institutions delivering them, even if not qualifying as “attacks” against them.
Distinction and constant care in military operations
Instead of “attacks”, some IHL provisions purport to protect civilians during “military operations”.Footnote 143 Specifically, IAC combatants may direct their military operations “only against military objectives”Footnote 144 and must take “constant care” in their conduct “to spare the civilian population”.Footnote 145 In both IACs and NIACs, civilians “enjoy general protection against the dangers arising from military operations”.Footnote 146 This requires the combatants not only to abstain from attacks against civilians and civilian objects but also to minimize “incidental losses” and to take reasonable precautions.Footnote 147 Together, these rules would sharply restrict the use of disinformation in armed conflict to the extent that it may be considered a military operation – though it is not clear that it would be.
While “military operations” are surely broader in ambit than “attacks”,Footnote 148 they are not defined consistently. In one section, the ICRC Commentary on the Additional Protocols distinguishes military operations from “ideological, political, or religious campaigns” based on the element of violence inherent to the former.Footnote 149 It further characterizes military operations as “all movements and acts related to hostilities that are undertaken by armed forces”.Footnote 150 In another section, however, the Commentary defines military operations as “movements of attack or defence by the armed forces in action”.Footnote 151
This erratic guidance leaves considerable ambiguity as to whether wartime disinformation might qualify as a military operation. It is possible that disinformation could be an act related to “hostilities”, which is interpreted broadly in IHL.Footnote 152 But it is unclear how “movements of attack” would differ from outright “attacks”, or when disinformation would be adjudged “violent” rather than “ideological” or “political”. Moreover, the “armed forces” may not be the sole entity engaged in disinformation operations. Even when they are involved, can they be said to be “in action” at that time? In short, many questions remain as to whether the IHL protections for civilians during “military operations” would apply to disinformation campaigns.
Criminal liability
Several of the criminal offences that might fit disinformation are premised on the occurrence of an attack. In both IACs and NIACs, the Rome Statute designates as war crimes “[i]ntentionally directing attacks” against civilians, civilian objects, humanitarian assistance personnel and objects, and hospitals.Footnote 153 Because war crimes are defined in reference to IHL, the meaning of “attacks” in these instances is the same as that given in AP I – i.e., “acts of violence against the adversary”.Footnote 154 As discussed above, this would probably exclude disinformation owing to the unlikeliness of the criminal conduct being attributed to the publisher.
For an act to qualify as a crime against humanity, it too must be committed in the course of a “widespread or systematic attack”.Footnote 155 However, the usage of “attack” differs here and “need not constitute a military attack”.Footnote 156 Instead, an attack in the context of crimes against humanity denotes “any mistreatment of the civilian population”Footnote 157 that “caus[es] physical or mental injury”,Footnote 158 including by “exerting pressure on the population to act in a particular manner”.Footnote 159 These broader glosses, drawn from the case law of international tribunals, might encompass disinformation that leads to civilian harm, including misrepresentations as to the quality or availability of essential goods and services. The disinformation in question would also have to be published in a “widespread” or “systematic” manner and “pursuant to a State or organizational policy”,Footnote 160 but these conditions are common features of coordinated media operations.
To the extent that disinformation causes or contributes to a health or humanitarian crisis, it might amount to the crime against humanity of “other inhumane acts … intentionally causing great suffering, or serious injury to body or to mental or physical health”.Footnote 161 Just as disinformation may interfere with the delivery of food, water or medical care, several international tribunals have found that similar deprivations in custodial settings constitute inhumane acts.Footnote 162 And if disinformation in an IAC were believed to have “wilfully imped[ed] relief supplies”, it might also be charged as the war crime of “starvation of civilians as a method of warfare”,Footnote 163 for which an attack is not a prerequisite.
Mental suffering
The final type of civilian harm owing to disinformation in armed conflict is mental suffering. Disinformation at any time can produce panic and distress, but all the more so during warfare. It is possible that these harms could be an unlawful attack against civilians or even a war crime or crime against humanity. It is indisputable that they violate the right to health, which includes mental health, though only insofar as IHRL jurisdiction applies.
Attacks and spreading terror
As noted above, acts that result in “injury or death” are “attacks” under IHL.Footnote 164 Importantly, “injury” may include “severe mental suffering”.Footnote 165 The meaning of “severe mental suffering” in IHL – including the level and types of harm contemplated and the evidentiary showing required – remains unsettled,Footnote 166 but criminal jurisprudence has provided a number of clues. International case law establishes that great suffering and serious mental harm “need not [be] permanent and irremediable”, but 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”.Footnote 167 Mental injuries might be caused by death threatsFootnote 168 or by otherwise subjecting a person to “intense fear, terror, [or] intimidation”,Footnote 169 and may be proven by medical records, victim statements or circumstantial evidence.Footnote 170 Some experts hold that post-traumatic stress disorder is sufficiently debilitating to be classified as severe mental suffering.Footnote 171 It is possible to imagine disinformation triggering similarly detrimental psychological reactions and therefore constituting a wrongful attack against civilians.
The mental harms brought about by disinformation might also evoke the IHL prohibition on “[a]cts or threats of violence the primary purpose of which is to spread terror among the civilian population”.Footnote 172 The similarity of this provision to the prevailing definition of attack (“acts of violence against the adversary”) is not coincidental – it is portrayed by the ICRC as merely an application of the antecedent prohibition on attacking civilians.Footnote 173 Indeed, it has in practice been primarily applied to incidents that unquestionably qualify as attacks, such as indiscriminate shelling or bombardment and physical assault or abuse.Footnote 174 To characterize disinformation as spreading terror, therefore, it must first be labelled an attack, or perhaps a threat of attack. Even if it is, however, it only conflicts with IHL if spreading terror is its primary purpose.Footnote 175 This may not be the case with disinformation, in which the principal objectives may also include punishing opponents, deflating morale or breeding confusion.
Criminal liability
Recalling the causation discussion above, it would likely be inappropriate to charge the perpetrators of a disinformation operation with offences involving killing or bodily harm. However, it may be possible to hold them to account for directly inflicting great suffering or serious mental injury upon civilians. Various forms of genocide, crimes against humanity and war crimes capture this unique form of harm. Because of the unique mens rea requirement (“intent to destroy, in whole or in part, a national, ethnical, racial, or religious group”), it is unlikely that disinformation would be prosecuted as the genocidal act of “[c]ausing serious bodily or mental harm to members of the group”.Footnote 176 However, disinformation might reasonably be charged as the crime against humanity of “other inhumane acts … causing great suffering, or serious injury to body or to mental or physical health”,Footnote 177 the IAC war crime of “wilfully causing great suffering, or serious injury to body or health”,Footnote 178 or the NIAC war crime of “cruel treatment”,Footnote 179 which is the infliction of “severe physical or mental pain or suffering.”Footnote 180
Right to health
Health in international law is understood in both its physical and mental dimensions. The International Covenant on Economic, Social, and Cultural Rights consecrates the “right of everyone to the enjoyment of the highest attainable standard of physical and mental health”.Footnote 181 Disinformation that damages mental health is a betrayal of this guarantee, though it must be considered which of the conflict actors bears this obligation with respect to the victim population.
Conclusion
Disinformation in armed conflict poses an ever-growing threat to civilians, who may be exposed to retaliatory violence, denied accurate information critical to their health or sustenance, or caused mental anguish by it. These harms have been overlooked in international law, in part due to the historical characterization of “misinformation” as a permissible ruse of war. This article has argued for the abandonment of that approach in light of its inconsistency with the contemporary nature of disinformation.
Even if this proposition is accepted, attempts to apply other rules of international law to disinformation encounter numerous grey areas and juridical impediments. First, while disinformation can be produced from afar, the applicability of IHRL and common Article 1 are both confined, respectively, to areas under the “control” or “authority” of conflict parties. Second, because the harms wrought by disinformation are often caused by indirect means, it is not likely to be viewed as an attack or other act of violenceFootnote 182 under IHL. Nor would it be a form of incitement, unless it also advocated violence or hostility. Third, while disinformation can inflict direct harms to mental health, these injuries are difficult to assess and are frequently neglected in IHL.Footnote 183 Fourth, disinformation may also frustrate the purpose of civilian objects, but not necessarily in a way that would amount to an attack. Finally, civilians and civilian objects directly participating in disinformation are unlikely to be deterred by the temporary stripping of their IHL protections, though they may be exposed to criminal liability.
The inadequacy of the legal response to disinformation in armed conflict, coupled with the gravity of the harms it may occasion, underscores the need for the articulation of clearer standards. In June 2021, the international community took an important first step with the release of the Oxford Statement on International Law Protections in Cyberspace,Footnote 184 which sets forth ten general principles, touching on sovereignty, incitement, human rights, criminal law, common Article 1, and general rules of IHL and international criminal law. This article has presented further details as to the specific application of these principles to disinformation during armed conflict, providing a theoretical foundation for further efforts in the field.
