Introduction
No part of life goes unimpeded by the devastation of war, but narratives of the negative impacts of war are dominated by physical harms – harms to life and to limb – and are often reduced to counting the number of individuals killed or injured. Yet, for every casualty statistic in conflict, there are countless more who have been harmed in a less visible but no less poignant way. This article draws attention to the overshadowed yet extensive mental harms of conflict; the one in five in conflict-affected populations who experience a mental health “disorder”Footnote 1 and the many more who experience mental harm beyond “disorder” categories, which are constructs centrally reflected in major clinical approaches to mental health diagnosis and treatment.Footnote 2 Many devastating mental harms occur when civilians are not intentionally attacked,Footnote 3 and these are thus referred to as incidental (or collateral) harms. Although the distinction between incidental and intended harms is key for determining the applicable legal norms, for those on the receiving end of the violence, the intention (or lack thereof) of the attacker may hold little relevance. This article undertakes a doctrinal analysis of international legal norms that do, or should, prevent mental harm in the conduct of hostilities, and proposes a critical analysis of the international humanitarian law (IHL) principle of proportionality in attack. Proportionality is the test through which the upper limit of permissible incidental civilian harm in attack is curated and which has thus attracted attention among the rules regulating the conduct of hostilities for its protective potential against incidental mental harm. In contributing to the ongoing debate on whether mental harm should be considered in jus in bello proportionality assessments, this article joins scholars such as Lieblich and Knuckey et al. in positing that it should.Footnote 4
A decolonial critique adds a new dimension to the debate and answers the compelling calls for cultivating a decolonial attitude to knowledge production.Footnote 5 In drawing on the work of decolonial scholars and the decolonizing global mental health agenda, this article remains cognisant of heterogeneous knowledges and the impact of colonialism and imperialism on knowledge creation,Footnote 6 and recognizes diverse understandings of mental harm. This decolonial lens reveals that piecemeal legal interpretations have encouraged the consideration of incidental mental harm in decisions of attack, but in a way that reproduces a Western-centric hierarchy between types of mental harms.Footnote 7 For example, some scholars, such as Lieblich, Schmitt and Highfill, and Knuckey et al., opine that the proportionality rule should be interpreted to include only certain mental health “disorders” like post-traumatic stress disorder (PTSD) or a traumatic brain injury (TBI).Footnote 8 However, such categorizations of mental harm rely on a Western-rooted conceptualization of mental health. The present article rejects this assumption of Western values and knowledge systems as universal or omnipotent, as it minimizes some lived experiences of mental harm through inappropriate contextualization or comparison.Footnote 9 Thus, it cautions against unduly narrow interpretations of the principle of proportionality that overlook much of the lived reality of mental harm and would block some individuals from receiving the protection of this key IHL principle. Instead, an alternative approach to reconciling proportionality in attack with incidental mental harm is proffered, grown in response to a critical, decolonial evaluation of the law and its interpretation and in light of the decolonizing global mental health agenda, but equally applicable beyond (post-)colonial contexts. At the core of this approach is the prioritization of the lived experiences of conflict-affected individuals – a factor relevant for the best practice implementation of IHL in all conflict situations. Two key features of this approach are suggested: (1) centring localized and culturally appropriate notions of mental harm, and (2) acknowledging socio-cultural and economic circumstances. Through its critical, decolonial exploration of the interpretation and implementation of the principle of proportionality, this article aligns with the United Nations (UN) Secretary-General’s call for a “more holistic approach [to civilian protection] that has meaning for all civilians affected by conflict”.Footnote 10
The article begins with an overview of the concept of mental harm through a decolonial lens before sharing reflections on the lived experiences of mental harm from attacks in armed conflict. It then summarizes the principle of proportionality in attack and its interpretation in existing literature and military policy regarding incidental mental harm. Next, the article critically evaluates mental harm within proportionality in attack through a decolonial lens, before concretely suggesting an alternative approach to implementing proportionality that seeks to respond to the decolonial critique. Finally, it articulates how this approach could be operationalized, anticipating potential critiques and exploring how the challenges of foreseeability, measurability, causal attribution and feasibility can be overcome.
Mental harm through a decolonial lens
Mental harm is used here expansively to cover challenges, impacts, injuries or impairments to mental, psychological, cognitive, behavioural and/or emotional well-being.Footnote 11 Engaging with the concept of mental harm should not be done without first acknowledging the Western ontological and epistemic hegemony embedded within the dominant conceptualization of mental health. This article seeks to recognize and respect non-Western ways of knowing and experiencing mental harm in conflict; it centres “world views of non-Western individuals” in order to reflexively unlearn and reimagine the assumptions and interpretations that order the world and construct knowledge.Footnote 12
The conceptual foundations of mental health are permeated with “European colonial and imperialist hegemony by privileging and reinforcing the perspectives of White, upper-middle class, Christian, heterosexual, male, and nondisabled individuals”.Footnote 13 The predominant approach to mental health mirrors a Western biomedical model of psychology and psychiatry that can be traced back to French philosopher René Descartes’ dichotomization of the mind and the body.Footnote 14 His Cartesian dualism – inherently a Western construct – initiated a global discourse (centred in European thought) that focuses on individual brain function and neurobiological explanations for mental ill-health.Footnote 15 This paradigm often juxtaposes alternative knowledges of mental health that hold validity in diverse socio-cultural contexts.Footnote 16 As Findlay summarizes, “[o]ne’s culture defines what it means to be well or unwell”;Footnote 17 thus, a Western-rooted construct of mental health cannot be assumed to be universally valid in all cultures. The themes of religion and spirituality, community collectivism, relationality and interdependence, mind-body-soul unification, connection to the environment, and socio-cultural circumstances are integral to the conceptualization of mental health across many diverse cultures, including some African,Footnote 18 AsianFootnote 19 and indigenous communities.Footnote 20 For example, in general, Muslims from Rakhine State in Myanmar do not share the notion of the mind–body divide; rather, they conceptualize mental distress through complex interrelationships between the “mind-soul”, the brain and the body.Footnote 21 Similarly, among some indigenous communities, mental distress is expressed through the notion of a “soul wound”.Footnote 22 Maori indigenous thought emphasizes the deep interconnectedness between humans and their environment (“healing the land to heal the people”), between the physical, mental, emotional and spiritual dimensions of existence, and between individual identity and community identity (whānau).Footnote 23 Horn observes that some cultures in the African continent also perceive mental well-being as a shared experience predicated on the notion of ubuntu (a collective self and mutual interdependence).Footnote 24 This does not align with the Western conceptualization of mental health, which is predicated on highly individualized understandings of the self.
Similarly, community, relationality and kinship are central to the understanding of mental health among Squamish people (indigenous people of southwestern British Columbia, Canada)Footnote 25 as well as in Filipino psychology through the concept of kapwa (meaning interconnectedness and a shared humanity).Footnote 26 Limenih et al. note that in some countries, cultural norms and social functioning patterns significantly influence how mental health is perceived, giving the example of how depression is comprehended within the broader socio-cultural context of the society rather than solely as a medical issue in some contexts.Footnote 27 The importance of socio-cultural circumstances to the understanding of mental health aligns with Fanon’s socio-diagnostic psychiatry (as interpreted by Mills), which posits that “one cannot understand psychological problems or distress outside of the conditions of oppression that lead to them”.Footnote 28 Evidently, conceptualizations of mental health are diverse, varying not only across but also within cultural settings.Footnote 29
Despite the significant variations in knowledge regarding mental health across the globe, the Western model has been systemically exported to “global South”Footnote 30 countries irrespective of socio-cultural context, including societies that have their own, alternative conceptualizations of mental health and where the Western ontological approach lacks coherence and validity.Footnote 31 This transplantation of Western-centric knowledge has been labelled as a form of “cultural imperialism”,Footnote 32 “modern colonisation”,Footnote 33 and “the colonization of minds”.Footnote 34 This ethnocentric knowledge on mental harm has been universalized through the imposition of standardized diagnostic categories for mental health “disorders”, which carry biases towards Western experiences of mental health.Footnote 35 The most widely used diagnostic manuals are the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM), currently in its fifth edition (DSM-5),Footnote 36 and the World Health Organization’s (WHO) International Classification of Diseases (ICD), currently in its 11th iteration (ICD-11).Footnote 37 Both these manuals perpetuate a Western-centric model of mental health “disorders” that relies on symptom checklists and a determination of “a clinically significant disturbance in an individual’s cognition, emotional regulation, or behaviour”.Footnote 38 Decolonial scholars such as Limenih et al. and Findlay have critiqued this transplantation of Western knowledge, arguing that it falsely assumes universal applicability or effectiveness, excludes significant variations in conceptualizations of mental health and culturally variable idioms of distress, and fails to adequately consider socio-cultural nuances in how mental health is experienced.Footnote 39 Consequently, these diagnostic manuals often fail to correspond with the important dimensions of community and relational experiences, the role of the soul, and connectedness to one’s environment that are focal in some cultures’ understanding of mental health.Footnote 40 For this reason, some decolonial scholars call for the assessment of mental harm to incorporate locally shaped and culturally recognizable expressions and experiences.Footnote 41
Encouragingly, DSM-5 and ICD-11 acknowledge some cultural variations, including – of particular relevance to this article – cultural factors that can heighten the likelihood of an attack causing mental harm. For example, ICD-11 recognizes that for some cultural groups, traumatic events affecting family members rather than the individual themselves, the desecration or destruction of religious symbols, or the inability to perform funeral rites could amount to a traumatic stressor for PTSD.Footnote 42 The latter is also referenced as a socio-cultural factor that can cause deviations in the onset of PTSD in DSM-5.Footnote 43 Nevertheless, the “disorder” categories themselves and their required diagnostic elements are static and impervious to cultural variations, despite the fact that they may be fundamentally unsuitable in some contexts. Due to the necessity of cultural relativity, some countries have integrated complementary diagnostic and alternative treatment systems within or alongside centralized ICD-11-based health systems.Footnote 44 Thus, if “disorder” categories based on these diagnostic manuals are used in isolation to inform the implementation of the proportionality rule (see the section “A Decolonial Critique of Existing Interpretations of Mental Harm within Proportionality” below), the protection of this rule will be limited to mental harm reflected within the manuals.
Further, while the inclusion of some cultural considerations in these manuals is promising, they remain minimal and under-developed. Localized research into the various ways in which mental harm is inflicted and presents itself in different socio-cultural contexts should be undertaken, not only to expand the cultural variation considerations within the manuals but also, crucially, to document patterns in experiences of mental harm that may fall outside of the manuals. “Disorder” categories have been criticized for obfuscating experiences of mental ill-health that fall below diagnostic cut-offs but that are not asymptomatic;Footnote 45 for example, research evidences that prognoses for individuals with sub-threshold depression (who have clinically relevant symptoms without meeting the full criteria for the disorder) are similar to those meeting diagnostic criteria.Footnote 46 To avoid the neglect of such experiences, some authors advocate for a dimension or gradient model of mental-ill health that assesses symptom severity on a continuum instead of subject to a threshold.Footnote 47
Not only can harm to mental health on a global scale not be adequately viewed through a Western lens, but that Western lens can also reinforce epistemic injustice.Footnote 48 DSM-5 and ICD-11 are culturally constituted objects not produced in a vacuum but informed by geopolitical dynamics, historical legacies, and the deeper ontological hegemony of the concept of mental health;Footnote 49 the need for concrete categorization is itself a “Eurocentric” ideology.Footnote 50 The use of diagnostic manuals that translate varied experiences of distress into “Western existential categories”Footnote 51 will necessarily dilute and eclipse local and indigenous knowledges and furthermore will preserve underlying power disparities and privileges that influence knowledge production.Footnote 52 It is also important to acknowledge the breadth of research evidencing the harmful impacts of taking a Western-centric approach to mental health practice or healing.Footnote 53 While not included here, these critiques remain pertinent as the discourse on mental harm shapes actions and thus the absorption of a Western conceptualization of mental harm into IHL norms could have a trickle-down effect into mental health interventions on the ground.Footnote 54
Mental harm from attacks in conflict
The vast and enduring mental harms caused in armed conflict have been widely evidenced, but they are often not disaggregated by the causes of the mental harm. For example, WHO’s latest estimate of the prevalence of mental health “disorders” in conflict-affected populations is 22.1%,Footnote 55 a 9.9% increase from the global mean estimate of 12.2% in 2019 – the same year WHO’s estimate was published.Footnote 56 Yet, this cannot all be attributed to attacks (defined under IHL as “acts of violence against the adversary”Footnote 57) – some mental harm is caused through other acts of violence that do not constitute an attack, such as sexual violence or torture,Footnote 58 while others may be caused by the circumstances of war that in some cases are not causally linked to an attack, such as lack of shelter, food or water, extreme financial hardship, unemployment, displacement or inability to access health care.Footnote 59 While it has been argued that protections against civilian harm in the conduct of hostilities extend beyond harm caused through attacks to the broader category of military operations, including by virtue of Article 57(1) of Additional Protocol I to the Geneva Conventions (AP I),Footnote 60 this article will focus on mental harm caused through an attack, in line with the codified principle of proportionality. Attacks are one of the gravest causes of mental harm in conflict, either directly or as a reverberating effect. Localized studies identify war-related stressors for mental harm, including air raids, artillery fire, the death or injury of loved ones, property destruction, witnessing the destruction of local infrastructure, and witnessing death or dead bodies.Footnote 61 The use of explosive weapons is evidenced as a particular direct cause of mental harm due to an acute stress reaction to the violence of the event.Footnote 62
Mental harm can also reverberate from physical harms caused by attacks. For example, Docherty notes that burn injuries can have long-term cognitive consequences due to the physiological state of shock caused by an extensive loss of fluids, and psychological and emotional consequences stemming from the burn incident itself and from the painful daily treatment, as well as lasting scars or disabilities.Footnote 63 Studies also evidence that individuals left disabled by war can experience profound psychological distress as a result, including PTSD, which was recorded at a prevalence of almost 30% among individuals with a conflict-related physical disability in Yemen.Footnote 64 Moreover, mental harm can be a cascading effect of socio-economic circumstances that result from an attack, such as family separation or economic hardship, which have been identified as mental harm triggers for children in conflict.Footnote 65 Socio-cultural and economic circumstances can also be underlying or concomitant determinants of mental harm that exacerbate the likelihood or severity of such harm materializing from exposure to a traumatic event, as documented in ICD-11 in relation to migrant populations.Footnote 66 Additionally, damage to the environment from an attack can cause mental harm, especially for communities who rely on ecosystem services or who have deep emotional and spiritual connections to the natural environment.Footnote 67 These direct and reverberating mental harms from attack can be severe, impeding quality of life, having community-wide impacts and altering the structures of societies, with long-term, lifelong and even intergenerational effects.Footnote 68 The UN Human Rights Committee recently asserted the intergenerational nature of mental harm caused by displacement, acknowledging the transmission of trauma across generations of Mayan people.Footnote 69
A limitation of the majority of available data on armed conflict-related mental harm is that it often relates only to specific categories of mental harm as stipulated in diagnostic manuals; for example, the aforementioned WHO estimate of 22.1% is a measure of five ICD-11 “disorder” categories: depression, anxiety, PTSD, bipolar disorder and schizophrenia.Footnote 70 Further examples include Karatzias et al.’s study of PTSD (as stipulated in ICD-11) of parents in UkraineFootnote 71 and McEwen et al.’s study evidencing a high prevalence of PTSD, oppositional defiant disorder, depression, and anxiety disorders (utilizing DSM-5 criteria) among Syrian refugee children in Lebanon.Footnote 72 The construction of these studies around “disorder” categories necessarily funnels experiences of mental harm in conflict through a Western frame; therefore, while they are indicative of the proliferation of profound mental harm, they inevitably omit mental harm that falls outside of or beneath these Western categorizations. In seeking – as this article does – not to be constrained by what Western epistemology renders as facts, attention should be drawn to literature that reports on mental harm outside of the terminology of a “disorder” and focuses instead on the manifestations of the distress. For example, Nicoll documents that for children in Gaza who have experienced and witnessed conflict, mental harm manifests through emotional and behavioural changes such as intense fear, bedwetting and reactive mutism.Footnote 73 Other studies observe mental harm manifesting through fatigue, intrusive memories, aggression, hyperactivity, avoidance, estrangement and self-isolation, sleep disturbance, moral injury and somatic symptoms.Footnote 74
This evidence of the reality of mental harm resulting from attack firmly establishes the significance of such harm and thus provides both the factual and moral basis on which this article argues for the progressive interpretation and implementation of the principle of proportionality in order to stem the infliction of mental harm during the conduct of hostilities.
International legal norms preventing mental harm in the conduct of hostilities
This article focuses on the principle of proportionality, which is at the centre of the debate on mental harm prevention in attack. However, the principle of proportionality is just one aspect of the legal framework that serves to prevent harm to mental health in the conduct of hostilities, which will be set out briefly below.Footnote 75
IHL enshrines a number of provisions that prohibit the intentional infliction of mental harm in the conduct of hostilities, including violence to mental well-being such as torture through mental pain or suffering (a jus cogens norm), cruel or inhuman treatment and outrages upon personal dignity, in particular humiliating or degrading treatment.Footnote 76 IHL further prohibits acts or threats with the primary purpose of spreading terror.Footnote 77 These provisions hold customary international law status and are applicable in both international and non-international armed conflicts.Footnote 78 International human rights law – applicable in both times of peace and conflict – provides additional protections, also prohibiting torture, cruel and degrading treatment that causes mental pain or suffering,Footnote 79 and genocide (a jus cogens norm) inflicted through “serious bodily or mental harm to members of the group”.Footnote 80 International criminal law bolsters these protections further by criminalizing torture, cruel and inhuman treatment and outrages upon dignityFootnote 81 as well as the act of genocide.Footnote 82 International law also contributes to the prevention of mental harm from attack through norms protecting civilian objects, essential services and critical infrastructure, cultural objects and places of worship, and the environment,Footnote 83 damage to which can be an underlying determinant of mental harm.
IHL provides protections against mental harm in the conduct of hostilities, even when it is inflicted incidentally. Three key provisions of IHL that enforce limits on acceptable incidental physical harm from attack have been interpreted by some to also apply to mental harm – the principles of precaution and proportionality in attack and the duty to take constant care. The principle of precaution in attack provides protections against injury to civilians,Footnote 84 even if it is not considered excessive in light of the principle of proportionality.Footnote 85 By nature of their mirrored language, the justifications for the reading of this provision to include considerations of incidental mental harm are interwoven with those for the principle of proportionality, discussed in detail below. The duty to take constant care to spare the civilian populationFootnote 86 has been interpreted broadly by Schmitt and Highfill to extend to all military operations, not just attacks, and to require “military personnel to avoid any harm to civilians, not just that which qualifies as incidental injury or collateral damage”.Footnote 87 Gillard expressly interprets this provision as requiring the consideration of mental health impacts.Footnote 88 While the principle of precautions in attack (if interpreted as such) and the duty to take constant care provide a degree of protection against incidental mental harm, even if fully complied with, unavoidable incidental mental harm will remain. Thus, if implemented in addition to these norms, the principle of proportionality could enhance the prevention of incidental mental harm further, acting as, according to Dorsey, “a final legal safeguard”.Footnote 89
The principle of proportionality
The principle of proportionality is at the centre of the debate on enhancing IHL’s protection against excessive incidental mental harm from attack. It is codified in Articles 51(5)(b) and 57(2) of AP I and is considered customary international law,Footnote 90 making it binding in both international and non-international armed conflicts. Further, breaches of proportionality during international armed conflicts constitute a grave breach of AP I as well as a war crime under the Rome Statute of the International Criminal Court (ICC).Footnote 91 The principle of proportionality prohibits “an attack which may be expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof, which would be excessive in relation to the concrete and direct military advantage anticipated”.Footnote 92 The three categories of harm listed in this provision (loss of civilian life, injury to civilians and damage to civilian objects), which it shares with the principle of precaution, represent the predominant understanding of civilian harm among legal scholarship.Footnote 93 However, it should be noted that outside of the confines of IHL, civilian harm is understood more broadly than these three categories and often includes mental harmFootnote 94 – for example, in the UN Secretary-General’s 2024 and 2025 reports on the protection of civilians.Footnote 95
Can “injury to civilians” be mental?
While the law does not provide a definition of injury for the purposes of understanding its meaning for the principle of proportionality, it is traditionally interpreted as purely physical injury.Footnote 96 As a result, excessive physical injury is prohibited while excessive mental injury is de facto neglected. This narrow interpretation significantly hampers the proportionality rule’s protection of civilians, yet, prima facie, there is nothing explicitly within the principle of proportionality that distinguishes between physical and mental harms or limits injury to the physical dimension. Further, as Bosi has shown, the drafters of IHL treaties were aware of mental harms as injuries of war, evidenced through the explicit inclusion of some mental health protections within other IHL provisions,Footnote 97 and significantly, within Article 51 of AP I itself.Footnote 98 Therefore, it is argued here that mental harm should not automatically be considered beyond the scope of the term “injury” where no express exclusion exists.
This position aligns with some experts who argue that there is no reason in principle for mental harm to be excluded from the notion of injury.Footnote 99 This view is held by Lieblich, a leading author on mental harm within the principle of proportionality, who posits that injury “could be reasonably understood to encompass also psychological harm”Footnote 100 and contends that such an interpretation is necessary “if IHL is to maintain its integrity”.Footnote 101 Similarly, Knuckey et al. argue that in principle mental harm should be accounted for in proportionality assessments, considering extensive scientific research into mental harm since the drafting of AP I.Footnote 102 An adjacent argument was raised by international experts on the conduct of hostilities attending a meeting co-organized by the International Committee of the Red Cross (ICRC) and Université Laval: they noted that IHL has been interpreted dynamically to consider new technology not known at the time of drafting, such as autonomous weapons, and thus an evolutive interpretation of IHL in line with the advanced understanding of the mental harms of war is equally justified.Footnote 103 A parallel can also be drawn between IHL’s concept of injury and its concept of the wounded and sick, as the latter has undergone an evolutionary interpretation and is now understood to encapsulate people with a mental health condition even though this was not expressly specified within the Geneva Conventions.Footnote 104
International criminal law jurisprudence supports the acceptance of mental harm within the notion of injury. For example, in Prosecutor v. Lubanga, the ICC Appeals Chamber expanded upon the definition of a victim of a crime under Rule 85(a) of the Rules of Procedure and Evidence, determining that the “harm” required to render an individual a victim includes “[m]aterial, physical, and psychological harm”.Footnote 105 Further, in Prosecutor v. Prilic et al., the Trial Chamber of the International Criminal Tribunal for the former Yugoslavia (ICTY) included the “significant psychological impact” of the destruction of the Old Bridge of Mostar in its analysis to determine that the attack was disproportionate.Footnote 106 The substance of proportionality was not engaged with by the judges when the appeal of the related conviction was upheld.
The Tallinn Manual 2.0 on the International Law Applicable to Cyber Operations (Tallinn Manual 2.0) is a persuasive source for incorporating mental harm into the IHL notion of injury, and thus the principle of proportionality. It interprets injury to include “severe mental suffering”, which the International Group of Experts agreed was in line with IHL’s underlying humanitarian and civilian protection agenda and further supported by analogy through IHL’s concrete provisions prohibiting intentional mental harm, indicating the drafters’ intention to include norms for mental health protection.Footnote 107 Yet, it is also important to note that the Tallinn Manual 2.0, in addition to some scholars,Footnote 108 does limit its interpretation to only “severe” mental harm; thus, while it is an authoritative source for encouraging considerations of incidental mental harm within the principle of proportionality, it is not without limitation as it could give rise to ambiguity around a legally relevant threshold of severity. While this article acknowledges the inability for injury to encapsulate all forms of mental harm, including minor impairments (as doing so would render the principle of proportionality inoperable), it argues that the evaluation of the severity of the mental harm ought not to be a precursor to a proportionality assessment, but a consideration within the proportionality assessment itself. The balancing act of proportionality will necessarily afford weight to different anticipated harms based on their gravity in relation to the concrete and direct military advantage anticipated – thus, severity does not need to be a pre-emptive determination. Ultimately, the wording of AP I does not limit the principle of proportionality to severe injury to civilians, be it interpreted as physical or mental.
Scholars also argue that the exclusion of mental harm from proportionality’s notion of injury might be contrary to AP I’s core objective of protecting civilians against the effects of hostilities.Footnote 109 As is well known, it is a fundamental rule of treaty interpretation that treaties must be interpreted in good faith in light of their object and purpose;Footnote 110 thus, a purposive interpretation of the principle of proportionality would prohibit instead of permit excessive incidental mental harm to civilians. Not following such an interpretation would undercut the humanity principle at the core of this legal norm, and of IHL more broadly.
Considerations when weighing mental harm in proportionality analyses
The proportionality assessment must weigh two values – civilian harm, including mental harm, and military advantage – against one another and determine if the anticipated civilian harm is excessive. This requires the consideration of multiple contextual factors, including the gravity and scope of the harm, the ability for the harm to be remedied in a timely fashion, the strategic importance of the military advantage, and the probability of both anticipated effects materializing.Footnote 111 Notably, the ICRC’s 1987 Commentary on the Additional Protocols holds that incidental civilian harm “should never be extensive”Footnote 112 – a condition that supersedes whether an attack is foreseen to comply with the principle of proportionality of not.
Jus in bello proportionality is assessed ex ante for a specific attack, and thus is observed by some armed actors as precluding the consideration of accumulated harm that results from multiple different attacks over time.Footnote 113 However, this article suggests that the fact that proportionality assessments are made in relation to specific attacks is not a barrier to the consideration of previous attacks as important contextual information to inform the analysis; rather, it argues that such considerations are required. The assessment of proportionality is inherently contextual, meaning that decisions of attack must be made considering all sources of information available to the individual at the time.Footnote 114 This argument is supported by Gillard, who asserts that belligerents must take into account the “specific context in which an attack will take place”, including damage caused by previous attacks, which may make further damage more significant.Footnote 115 Dorsey further emphasizes that proportionality assessments demand a context-specific balancing exercise of qualitative considerations and ethical nuances.Footnote 116 Contextual considerations will be particularly salient for weighing mental harm in proportionality analyses, as mental harm can be exacerbated through repeated attacks and attacks can compound with underlying socio-cultural and economic determinants to heighten the risk of mental harm materializing (see the section on “Acknowledging Socio-Cultural and Economic Circumstances” below). Diamond and van Benthem argue that factoring context and specific vulnerabilities of a population into proportionality assessments could enable those assessments to capture mental harm originating from multiple isolated or entwined sources, including previous attacks, displacement and malnourishment.Footnote 117
In an alternative approach, Lubell and Cohen call for a “strategic proportionality” assessment – a higher-level assessment of proportionality that is to be conducted throughout the conflict and which balances “the overall harm against the strategic objectives”.Footnote 118 They posit that this would enable the cumulative effects of mental harm to be considered in decisions of attack, even when the harms do not have a singular cause but result from prolonged exposure to war.Footnote 119 This approach may enable anticipated incidental mental harm to have a greater influence on decisions of attack, but considerations of cumulative harm would be limited to that which is causally linked to a single armed actor, leaving a loophole where mental harm results through exposure to attacks by multiple parties and omitting the consideration of other contextual factors and underlying determinants that this article proposes should be taken into account. Additionally, while Lubell and Cohen’s approach allows for cumulative harm to be considered, it is to be balanced against the overall strategic aim of the use of force – thus, the scope has been widened on both sides of the proportionality equation.Footnote 120 It is hard to conceive of many circumstances in which expected overall incidental mental harm would be rendered disproportionate against the alleged high-level strategic objectives of the conflict.
Military policy on incidental mental harm
Military policy largely mirrors the language of AP I with regard to the principle of proportionality, but injury to civilians is also captured in various military manuals through alternative phrasings of “suffering”, “harm”, “harmful effect”, “affecting the civilian population” and “damage to persons”.Footnote 121 Whether the broad formulation of these policies is intended to permit them to be interpreted to include mental harm can only be speculated. Other authors’ assessments of State practice indicate that militaries generally do not factor mental harm into their considerations of proportionality of attack;Footnote 122 however, some military policies can also be drawn upon to illustrate degrees of receptiveness to considering mental harm in judgements of proportionality. For example, the US Department of Defense’s (DoD) 2023 instruction on Civilian Harm Mitigation and Response (CHMR DoD-I)Footnote 123 mandates consideration of a narrow scope of civilian harm which it defines as “civilian casualties and damage to or destruction of civilian objects”.Footnote 124 A report by the US Government Accountability Office acknowledges that this civilian harm definition sought to exclude “second and third order effects, such as psychological effects”.Footnote 125 However, the CHMR DoD-I also appears to partially extend its civilian harm mitigation efforts to a broader category by requiring consideration of “other adverse effects on the civilian population”, albeit only requiring consideration of these effects “to the extent practicable”.Footnote 126 Although the instruction does not elucidate on what other adverse effects it intends to capture, the wording is broad and thus could be interpreted to cover those harms that it excluded from its narrow conceptualization of civilian harm, notably mental harm. Nevertheless, even if the CHMR DoD-I can be interpreted to require some consideration of mental harm, it still maintains a hierarchy that prioritizes physical harms for which consideration is mandatory while “other adverse effects” – potentially including mental harm – are only required to be considered where practicable.
A second example is from North Atlantic Treaty Organization (NATO) policy, which is more explicit and progressive in its extension of IHL to incidental mental harms. In its incorporation of the principles of proportionality and precaution, NATO’s 2021 revised joint targeting doctrine requires analyses of “expected collateral damage” to include estimations of “effects in the virtual and cognitive dimensions”,Footnote 127 with “cognitive dimensions” expanded on as “psychological/behavioural effects”.Footnote 128 However, the doctrine does acknowledge that “the risk estimate for effects in these two dimensions [virtual and cognitive] may not achieve the same level of prediction as the physical one”, and it recognizes that “the integration of cognitive effects into NATO Joint Targeting is still in its infancy”.Footnote 129 Although US and NATO policies can be viewed as somewhat receptive to the mitigation of excessive incidental mental harm, this analysis is based on policy language and whether mental harm is, or could be, conceptually captured by the policies, rather than on observations of military practice.
A decolonial critique of existing interpretations of mental harm within proportionality
Having presented the current literature and policy on proportionality and incidental mental harm, this article will now evaluate mental harm within proportionality in attack through a decolonial lens, and (in the subsequent sections) explore how this critique could be overcome through suggesting what an alternative approach might look like and how it could be operationalized.
The growing literature encouraging the inclusion of incidental mental harm in the principle of proportionality gives hope towards better policy and strategic protections for mental health in conflict; however, the present article engages with a pressing challenge that has not yet been acknowledged or addressed in this debate. In interpreting the principle of proportionality to include mental harm, some scholars impose narrow parameters on the mental harm that they seek to include, and this enforces a hierarchy within types of mental harm that privileges and prioritizes Western experiences and knowledge. Two problematic conceptualizations are identified below: (1) the separation of physical and psychological harm, and (2) the utilization of Western-centric “disorder” categories. It is argued here that perpetuating these Western-centric conceptualizations of mental health diminishes other forms of knowledge and, on a practical level, could, in certain contexts, lead to skewed interpretations of proportionality that do not sufficiently consider the foreseeable mental health impacts of an attack.
For example, in their paper on whether the causation of TBIs to civilians should, under IHL, be considered in decisions of attack, Schmitt and Highfill interpret injury for the purposes of the principle of proportionality to extend, in theory, to considerations of incidental mental harm that result from a TBI.Footnote 130 However, they appear to exclude any other manifestations of mental harm from the reach of the principle of proportionality, asserting that, in light of the state of the law in 2018, the term “injury” “could not be interpreted to encompass purely psychological harm unrelated to TBI”.Footnote 131
A similar conceptual separation between biological harm to the brain (requiring medical treatment) and purely psychological harm (treated through psychotherapy or psychological and behavioural interventions) was suggested within the International Law Association Study Group on the Conduct of Hostilities in the 21st Century (ILA Study Group), with some members arguing that the latter should be excluded from the principle of proportionality.Footnote 132 However, this distinction appears convoluted, is not a distinction made within the principle of proportionality itself, and fails to acknowledge that some manifestations of mental harm unrelated to a brain injury can still cause biological changes to the brain and result in physical symptoms.Footnote 133
Lieblich also restricts his interpretation of proportionality considerations to certain experiences of mental harm, although in a different way than Schmitt and Highfill.Footnote 134 He writes that “[w]hen we discuss mental harm, we naturally refer to the most serious, well studied forms of such harm, such as post-traumatic stress disorder”.Footnote 135 Knuckey et al. similarly limit their discussion of incidental mental harms exclusively to PTSD, although they do acknowledge cross-cultural variations of the “disorder” and the impact of confounding socio-economic variables.Footnote 136 Solomon notes that “it is wrong to only associate civilian mental harm with PTSD”, but explains that PTSD is used by legal scholars due to its ability to translate mental harm into the legal sphere.Footnote 137
Crucially, these authors’ interpretations of the principle of proportionality and its required mental harm considerations largely rest on the Western-centric “disorder” categories of TBI and PTSD, as coded and classified in ICD-11 and DSM-5. The aforementioned concerns over the cross-cultural validity of these diagnostic manuals should not be overlooked by those seeking to envelop “disorder” categories into the application of the principle of proportionality.Footnote 138 While these authors encourage greater consideration of incidental mental harm in decisions of attack than is currently undertaken in practice, they limit incidental mental harm considerations to standardized “disorder” categories that depict only certain forms of mental suffering as legible,Footnote 139 and their approaches would therefore omit the broader landscape of lived experiences of mental harm in conflict that do not fit neatly into “disorder” categories from the protective scope of the principle of proportionality.Footnote 140 Such an interpretation does not align with WHO’s definition of mental health as “more than the absence of mental disorders”.Footnote 141 The International Criminal Tribunal for Rwanda (ICTR) took a different approach to interpreting the necessary severity of serious mental harm for the crime of genocide: it held that the mental harm must include “more than minor or temporary impairment of mental faculties” but need not be “permanent or irremediable”.Footnote 142 Following this approach to incidental mental harm considerations in decision of attack would permit the necessary flexibility to embrace cultural diversity in lived experiences of mental harm, rather than using disorder categories as thresholds of severity for mental harm to be considered and ultimately mitigated.
The limitations of forcing lived experiences of mental harm through rigid “disorder” categories can be illustrated through the example of PTSD, a diagnosis constructed partly in response to the mental suffering of US veterans returning from the Vietnam War.Footnote 143 Scientific research into PTSD is mostly consumed by the experience of trauma by Western soldiers,Footnote 144 but it goes without saying that the civilian and combatant experiences will be very different. Patel and Hall find that PTSD symptoms are “culturally malleable” and observe that for Cambodian refugees the local idiom of “thinking a lot” was a more prominent trauma reaction than any of the symptoms prescribed for PTSD in DSM-5.Footnote 145 Research has shown that even in socio-cultural contexts where PTSD is a commonly accepted diagnostic construct and its symptomology holds validity, its clinical utility can be limited – such as in Afghanistan, where compared to culturally specific and indigenous expressions of distress, PTSD has a lesser substantive impact on people’s daily functioning.Footnote 146 The impermeability of the construct of PTSD to cultural nuances has led some mental health experts to reject its pertinence to their population. For example, the chair of the Palestinian Ministry of Health’s Mental Health Unit, Dr Samah Jabr, critiques the clinical definition of PTSD for failing to fully capture the experience or reactions of Palestinians experiencing collective and continuous trauma.Footnote 147 Horn critiques the highly medicalized discourse of PTSD as a “Northern orthodoxy” that is “predicated on a privileged idea that the world is fair and just in the first place”;Footnote 148 indeed, cultural relativists would challenge the applicability of PTSD anywhere outside of the culture where the concept was constructed.Footnote 149 Confining mental harm considerations for proportionality analyses to “disorder” categories like PTSD could result in inaccurate assessments of the proportionality of an attack that omit culturally nuanced trauma responses.
In addition to omitting experiences that fall outside of Western-constructed parameters, relying on “disorder” categories when anticipating incidental mental harm from an attack requires a distinction between different types of mental harm. Not only does the codified law for the principle of proportionality not require distinctions between types of civilian injuries, but current practice and collateral damage estimate (CDE) methodologies also do not categorize or typologize foreseeable incidental physical harms.Footnote 150 Thus, through analogy, it can be argued that distinction between different types of mental harm is not required in law either.
Responding to the decolonial critique: A culturally and contextually sensitive approach to mental harm within proportionality
A decolonial evaluation illustrates the need for a localized, socio-culturally sensitive interpretation of mental harm within the principle of proportionality. This section will explore practical suggestions for interpreting and implementing the principle of proportionality in order to overcome concerns highlighted through a decolonial lens. While not reflecting current prevailing practice, this approach would respond (at least in part) to the above decolonial critique, but it also holds value beyond (post-)colonial contexts, representing the best practice of centring lived experiences in the principle of proportionality, applicable in all conflict situations where IHL applies. The following key features are explored below, alongside suggestions for how they could be concretely applied to the principle of proportionality: (1) centring localized and culturally appropriate notions of mental harm, and (2) acknowledging socio-cultural and economic circumstances.
Centring localized and culturally appropriate notions of mental harm
Proportionality could be interpreted in a manner that recognizes epistemological diversity on mental health and the need for cultural humility,Footnote 151 for example by ensuring that diverse cultural understandings of mental harm are not ignored. This article has identified that community collectivist values and interdependence, mind–body–soul unification and connection to the environment are key features of mental well-being in some cultures (see the above section on “Mental Harm through a Decolonial Lens”). Cultural nuances in the conceptualization of mental health and mental harm could be incorporated into research that might inform proportionality analyses, ensuring that relevant lived experiences and perspectives shape the future implementation of proportionality in attack. Encouragingly, progression towards this culturally nuanced approach is nascent. NATO’s 2021 revised joint targeting doctrine acknowledges how, in some cultures, mental health is indivisible from community interconnectivity or environmental connection. In the context of “CDE for effects in the virtual and cognitive dimensions”, it states that “commanders and their staffs should manage the risk by making efforts to understand the human environment”, within which it includes “how all humans interact with their environment, especially with each other”.Footnote 152
Research could also embrace local languages and idioms of distress relevant to particular conflict contexts, such as the term waushanti (meaning sad or restless mind), used by conflict-affected communities in Rakhine State in Myanmar,Footnote 153 or the Afghani concept of jigar khun (meaning a state of dysphoria or melancholy), used widely to describe war-related experiences.Footnote 154 The prominence of local idioms of distress in an individual’s trauma response is emphasized in parts of ICD-11,Footnote 155 representing a positive advance toward cultural adaptation of “disorders” within diagnostic manuals. Embracing local knowledge and beliefs on mental harm in these suggested ways would respond to the aforementioned decolonial critique that rejects the Western framing of mental health in favour of “home-grown” culturally authentic and contextually appropriate conceptualizations and frameworks of mental suffering that are grounded in the realities of affected communities and “privileg[e] the knowledge of those with lived experience of distress”.Footnote 156 Centrally, making localized and culturally nuanced information on mental harm accessible to armed actors would facilitate its consideration in future proportionality assessments, ensuring that incidental mental harm considerations are shaped by relevant lived experiences and not limited to those which align with Western notions of mental harm.
Acknowledging socio-cultural and economic circumstances
Another key feature in identifying mental harm for proportionality in attack compliance, proposed by this article, is the consideration of socio-cultural and economic circumstances of the war-affected community at risk of mental harm. Two complementary avenues through which this could be improved are explored here. Firstly, some attacks which cause or exacerbate underlying socio-cultural and economic determinants of mental harm (such as extreme financial hardship, food insecurity, displacement and a lack of access to health careFootnote 157) could be captured within the remit of the principle of proportionality through the notion of reverberating effects. The prevailing view among legal scholars is that IHL requires reverberating harms to be considered in proportionality analyses, provided they are an expected consequence of the attack;Footnote 158 however, there remains some disagreement as to which reverberating harms must be considered. For example, while State policy and practice appear to interpret reverberating harms as necessary considerations in proportionality assessments, the material scope of the reverberating incidental harm appears limited in some policies, such as the US Law of War Manual’s exclusion of some reverberating economic harms from determinations of proportionality.Footnote 159 Scholars have called for incorporating more detailed information regarding characteristics and vulnerabilities of civilian infrastructure and services for the purpose of improving reverberating harm mitigation,Footnote 160 thus illustrating a willingness to place greater reliance on socio-cultural and economic circumstances within proportionality assessments. However, the UN Secretary-General has observed that the reverberating effects of conflict on “individual mental and societal trauma” are not factored into proportionality analyses by some parties to conflict.Footnote 161 Ongoing debate regarding the scope of this obligation and the point at which a reverberating effect becomes confidently predictable enough to be consideredFootnote 162 could be an obstacle to the uptake of this approach.
Secondly, socio-cultural and economic circumstances that may heighten the risk of mental harm being inflicted, should be considered when determining whether an attack is proportionate. As argued above, the assessment of proportionality is inherently contextual, and decisions of attack must be made considering all sources of information available to the individual at the time.Footnote 163 Gillard gives the example of the capacity of local health facilities as a contextual factor that should be considered in proportionality assessments.Footnote 164 Some military policies do consider cultural factors when weighing up the proportionality of an attack: for example, the US Law of War Manual affords “greater consideration” to incidental damage of cultural property than to ordinary property.Footnote 165 Aforementioned cultural factors identified in ICD-11 and DSM-5 that can heighten the risk of PTSD – such as religious beliefs that can result in mental harm when religious symbols are destroyed – serve as an example of how cultural context can inform proportionality assessments, even within the confines of “disorder”-based approaches.
In order to centre localized and culturally appropriate notions of mental harm, the implementation of proportionality in attack should include the consideration of further contextual and circumstantial factors, such as the accessibility of mental health-care services, mental harm caused by previous attacks or natural disasters and socio-cultural and economic factors that indicate likely mental harm triggers. As an illustration, attacks that may have been deemed proportionate in the first week of a conflict may fail the proportionality test after two years of the community being exposed to numerous mental harm triggers, because such contextual considerations – which indicate a higher risk of mental harm – would, through the utilization of this article’s approach, be absorbed into the proportionality analyses provided they are reasonably foreseeable. Notably, this proffered contextual approach would not limit armed actors to considerations of how their own previous military actions have affected the community; rather, it would require them to consider how all existing mental health impacts or triggers for which information is available – regardless of who or what inflicted them (such as another armed actor or a natural disaster) – could compound the foreseeable mental harm of their attack.
Solomon and Bayer show progression towards greater contextualization of jus in bello proportionality. They evidence that socio-economic circumstances can shape the materialization and intensity of mental harm resulting from rocket attacks, which supports their argument for a contextual approach to incidental mental harm, with proportionality assessments taken on a case-by-case basis without predetermined assumptions.Footnote 166 Solomon and Bayer criticize the horizontal approach to mental harm in IHL that falsely premises that war is experienced equally by all.Footnote 167 Their position takes crucial steps towards a holistic approach to incidental mental harm that aligns with decolonial calls for mental harm not to be siloed from socio-cultural and economic circumstances and determinants.
These two key features embrace alternative knowledges that view socio-cultural and economic circumstances as crucial to understandings of mental harm and would account for the intersecting, compounding and cumulative nature of conflict exposure, the conflict’s longevity and daily stressors. Embracing these features in the implementation of the principle of proportionality would thus respond to the UN Secretary-General’s call for an approach that “takes into account the complexity and cumulative nature of the full range of civilian harm”.Footnote 168
Operationalizing a culturally and contextually sensitive approach to mental harm within proportionality
How the above-stipulated approach to incidental mental harm may shape the implementation of the principle of proportionality will vary across different conflict contexts, different identities of affected communities, and different military advantages pursued. The absorption of localized knowledge on the manifestation of mental harm would enable proportionality analyses to consider a more truthful representation of the types and gravity of incidental mental harm likely to result from a particular attack. Such insight may add weight to the civilian protection side of the proportionality balance and could, theoretically, shift the balance of proportionality; for example, a broader interpretation of mental harm that goes beyond Western “disorder” categories could capture a greater proportion of foreseeable mental harm resulting from a military attack. Considerations could include collective experiences of mental harm, mental harm that falls just below or outside of current “disorder” categories, or mental harm emanating from the destruction of or damage to social relationships, cultural heritage or the environment. Depending on the anticipated military advantage, this could require, in practice, alterations to the means and methods of attack in order to mitigate or prevent the foreseeable mental harm.
The commonly cited operational challenges to considering mental harm in jus in bello proportionality are foreseeability, measurability, causal attribution and feasibility,Footnote 169 although some authors have made strong rebuttals against these perceived barriers.Footnote 170 This section briefly summarizes these concerns and their rebuttals and undertakes an initial analysis of how the proffered culturally and contextually sensitive approach to mental harm may intersect with, exacerbate or ease existing operational challenges, and how those challenges could be surmounted. Although it is key to be aware of the practical hurdles of embracing this approach, they should not be used to justify the continued ignorance of mental harm in decisions of attack, or indeed the perpetuation of a Western-centric approach to incidental mental harm. Rather, these challenges highlight areas where further research is needed to ameliorate and overcome these operational concerns and build confidence in the utility of a culturally and contextually sensitive approach for improving civilian protection in war.
Foreseeability
The codified test of the principle of proportionality requires incidental civilian injury to be considered in decisions of attack, if the attack “may be expected to cause” the injury. This means that those involved in determining the proportionality of an attack would not be expected to predict all possible harms (mental or physical), only those reasonably foreseeable by a person who is reasonably well informed.Footnote 171 Schmitt and Highfill observe that reasonable foreseeability is based on rational inferences and probabilities but excludes conjecture and speculation.Footnote 172 Scholars have raised concerns as to the foreseeability of incidental mental harm by armed actors, military strategists and policy-makers due to its heightened subjectivity in comparison to physical harm.Footnote 173 The subjectivity critique is emboldened when mental harm is viewed through a decolonial lens, as it invites consideration of socio-cultural and economic factors and determinants. Nevertheless, it does not make foreseeing incidental mental harm an impossibility; indeed, the principle of proportionality already deals with highly subjective harms. Firstly, many manifestations of physical harm are non-homogeneous (for example, age, disability, gender and many other factors can impact the infliction of physical harm) and thus can be difficult or impossible to predict, such as determining who might be struck, and where on their body they might be struck, by shrapnel which could lead to diverse harms ranging from minor to fatal.Footnote 174
Secondly, as noted above, the incorporation of reverberating harms in proportionality assessments is increasingly accepted in State practice,Footnote 175 including, notably, the US and UK military manuals, which both give examples of reverberating harms that ought to be considered in decisions of attack.Footnote 176Anticipating reverberating harm can similarly be highly complex and subjective to local circumstances and socio-cultural and economic factors. Like mental harm, reverberating effects can only be considered insofar as they are reasonably foreseeable. The fact that some reverberating harms are not reasonably foreseeable is not a barrier to the determination that the law obliges expected reverberating harms to be taken into account in proportionality assessments, and nor should that be the case for mental harm. While foreseeability may limit some mental harms from being considered, it is not a valid justification per se for excluding mental harm as a category of injury from the principle of proportionality. Concerns regarding the foreseeability of mental harm (especially when socio-cultural and economic factors are engaged with) could be further eased by using Talhami and Zeitoun’s “precautionary approach” to foreseeing reverberating harm. This approach “assumes causal links and chains” within military operational planning processes to address the inevitable uncertainty in predicting the harm that may result from an attack.Footnote 177 It would also aid issues of causal attribution (discussed below).
Not all socio-cultural and economic factors or individual circumstances that could enhance the likelihood of mental harm materializing from an attack will be reasonably accessible to decision-makers, but some contextual information will be obtainable or even publicly available and therefore should be considered when anticipating incidental mental harm. Indeed, information already collected in attack planning in order to comply with IHL norms on the conduct of hostilities may be useful for foreseeing mental harm. Information on foreseen civilian injuries could help predict mental harm that may result, for example, in war-disabled individuals, while estimates of foreseen civilian deaths could inform mental harm predictions through the impacts that these deaths may have on family units and broader communities. Further, the identification of civilian objects within the operating environment, including no-strike entities such as cultural, religious or historical sites,Footnote 178 can be used to anticipate mental harm that may result from attacks near to these sites. Population density estimates for civilians in the operating area at specific times,Footnote 179 as well as local demographics,Footnote 180 could prove particularly useful, especially for identifying children, who have a heightened susceptibility to mental harm.Footnote 181 Finally, the foreseeability of mental harm could be improved through the utilization of information regarding underlying social-cultural determinants of mental harm (such as the culture, history, welfare and social structure of the civilian population within the operating environment) that is collected by some armed actors as part of their efforts to understand the “human environment”,Footnote 182 as well as mappings of crucial civilian services such as water and electricity.Footnote 183
More empirical data on how contextual factors alter the infliction of mental harm from an attack could also be collected and made available to decision-makers to enhance the foreseeability of mental harm. Pivotal to this will be battle damage assessments, including manifestations of mental harm from previous attacks that can inform future proportionality analyses, both to improve foreseeability and to ensure that considerations of compounding harm are included. Rigorous research including granular anthropological studies that evidence localized experiences of mental harm and how diverse socio-cultural factors intersect with the manifestation of mental harm could be undertaken. In addition, large-scale quantitative studies that map data on mental harm resulting from previous attacks to information about conflict typology, location, and means and methods of warfare could be used to establish various mental harm probabilities. Such research could shape mental harm prediction models that can be embedded into existing incidental civilian harm assessment algorithms or calculations (that must accompany and not replace qualitative human reasoning),Footnote 184 as suggested by Lieblich in relation to PTSD.Footnote 185 Crucially, this further research should not be contoured around rigid mental health “disorder” categories to ensure that it does not exclude aforementioned experiences of mental harm which do not align with those categories.
Foreseeability could also be strengthened by drawing robust inferences with accepted legal categories of intentional mental harm, such as psychological torture and acts of terror. For example, depending on the intensity of the suffering inflicted, sleep deprivation can be recognized as either psychological torture or inhuman treatment under international human rights law,Footnote 186 and thus it is foreseeable that military attacks which will result in persistent sleep disturbance could cause mental harm.
Measurability
The perceived intangible nature of mental harm is commonly quoted as a barrier to its measurability and thus a justification for its exclusion from proportionality analyses. This has been convincingly rebutted by Lieblich and others, who observe the intangibility of some physical harms and the fact that mental harm is sufficiently tangible to be considered in other international and domestic legal provisions, including parallels drawn from tort law.Footnote 187 Other concerns over the measurability of anticipated mental harm rest on the challenge of assigning weight to it to balance it against a concrete and direct anticipated military advantage. Such concerns are rebutted by Gillard, who notes the equal difficulty of assigning weight to physical harm.Footnote 188 The weighing and balancing inherent within the principle of proportionality is not an exact science and will always be flawed since the values it seeks to compare (military advantage and civilian harm) are not directly commensurable. The opacity in weighing anticipated mental harm is therefore not unique to mental harm but is linked to the principle of proportionality itself. Notably, the ICRC’s 1987 Commentary on the Additional Protocols asserts that when there is “hesitation” over whether an attack would be disproportionate, “the interests of the civilian population should prevail”.Footnote 189
A culturally and contextually sensitive approach to mental harm raises additional challenges for measurability due to the diverging tension between quantification and contextualization. Quantitative studies (as called for above) will be key to informing prediction models for anticipating mental harm, but the quantification of harm has been criticized by Glasman, among others, for its decontextualization, individualization and erasure of collective or relational social experiencesFootnote 190 – crucial considerations highlighted in this article. Other critics also observe how the construction of quantitative knowledge is shaped by conceptual and epistemological particularities and the interpretive creation of categories, as a result of which it does not necessarily entirely reflect reality.Footnote 191 Thus, it is argued here, the quantification of mental harm in research intended to inform decisions of proportionality must be accompanied by localized qualitative information. To ensure that measuring mental harm does not come at the expense of contextualizing lived experiences, policy-makers, military strategists and armed actors can rely on the wealth of research that already exists regarding localized approaches to mental and emotional well-being. Of particular importance are the likes of Rasmussen et al.’s systematic review of diverse cultural concepts and idioms of post-traumatic stress that sit outside of standardized “disorder” categories.Footnote 192 Ultimately, a balance must be struck between the contextualization and cultural sensitivity of mental harm on the one hand, and on the other, the need for robust data to inform prediction models and enable confidence in the anticipation of mental harm.
Causal attribution
As the principle of proportionality seeks to capture harms that result from a single attack, it requires a foreseen causal relationship between the planned attack and the anticipated incidental harm. The causation requirement is an ex ante assessment of whether an attack “may be expected to cause” harm, not an assessment of factual causation; hence, anticipation that mental harm would result from an attack, and the expectation that it would not occur but for the attack, is sufficient evidence of causation.Footnote 193 Nevertheless, the issue of causal attribution is a frequently cited obstacle to considering incidental mental harm in proportionality assessments, often justified on the basis that mental harm does not always have a clear, sole cause:Footnote 194 for example, when prolonged exposure to hostilities erodes mental well-being incrementally over time, it can be hard to pinpoint one attack as the cause or rule out potential intervening causes.Footnote 195 Solomon notes that this issue is subverted by scholars who limit their assessment of incidental mental harm to PTSD, due to its diagnosis requiring a triggering traumatic eventFootnote 196 – something that is relatively easy to anticipate in an active conflict.Footnote 197 However, this article has suggested that anticipating the incidental mental harm of an attack should not rely exclusively on rigid “disorder” categories, and thus it cannot benefit from the causal clarity of PTSD, and nor can other “disorders” that do not contain clear criteria on the causal event triggering the harm. Two central features of the principle of proportionality are relied on here to appease concerns regarding causation: its acceptance of indirect causation and its inherently contextual nature.
First, the principle of proportionality allows for indirect causation, which in turn may accommodate some of the often convoluted causes of mental harm. As codified in AP I, the principle contains no requirement for the anticipated harm to be caused directly by the attack (unlike the anticipated military advantage, which does need to be direct). Therefore, causation can be satisfied through a chain of anticipated events, with no limit on the number of causal steps indicated in AP I.Footnote 198 This reverberating dynamic is particularly amenable to a culturally and contextually sensitive approach to mental harm as it allows for harm to be causally attributable to an attack even if it is not geographically or temporally proximate,Footnote 199 such as mental harm caused indirectly through the infliction of underlying determinants (see the above section on “Acknowledging Socio-Cultural and Economic Circumstances”). Here proportionality assessments could be aided by further research, including the aforementioned large-scale quantitative studies exploring patterns of mental harm in previous attacks, with the added focus of mapping the specific causal chains and relationships to a sufficient granularity necessary to inform prediction models; this should include longitudinal studies to capture experiences of mental harm that may not be temporally proximate to the causal attack. Second, the highly contextual nature of the principle of proportionality (see the above section on “Considerations When Weighing Mental Harm in Proportionality Analyses”) makes it easier to satisfy causation for incidental mental harm as it requires armed actors – as argued in this article – to consider other underlying social, cultural or economic factors or vulnerabilities that elevate the risk of their attack causing mental harm.
Feasibility
Criticisms of including incidental mental harm in proportionality assessments often centre on feasibility, such as the critique that it would warrant members of armed forces or groups to possess unrealistic psychiatric expertise.Footnote 200 However, in operationalizing incidental mental harm considerations in proportionality analyses, this article has proposed further research that can be utilized in prediction models. These models could be incorporated into existing incidental civilian harm assessment algorithms or calculations, thus negating expectations of medical expertise for armed actors with sophisticated civilian harm estimation technology and for pre-planned attacks. Decisions made in the heat of battle will benefit less from this.
It could also be argued that a culturally and contextually sensitive approach would invite a flood of minor mental harm considerations into proportionality analyses that could paralyze military activities since targeting decisions often need to be made rapidly during conflict. However, the act of balancing to determine proportionality will necessarily set aside minor mental harms in the face of concrete military advantages, just as minor physical harms are, in most conceivable scenarios, unlikely to alter armed actors’ planned attacks. In requiring the consideration of incidental mental harm, it does not follow that said harm must be mitigated or prevented, but rather that operational adjustments towards avoiding or in any event minimizing the harm should be explored.Footnote 201 Nevertheless, non-mitigatable mental harms, including those that fall outside of or beneath Western “disorder” categories, ought to still be acknowledged as lived experiences of conflict, and not ignored.
Criticisms of incorporating mental harm into proportionality assessments often hyperbolize the uncertainty surrounding mental harm and juxtapose it to the predictability of physical harm without acknowledging that jus in bello proportionality is, by design, saturated with uncertainty and “famously vague”.Footnote 202 Neither side of a proportionality equation can be predicted without a margin of error. Lieblich provides a crucial reminder that proportionality is “intrinsically woven with the notion of ‘feasibility’”;Footnote 203 thus, reasonable foreseeability of incidental mental harm, in light of what is feasible, is all that is required.
Finally, arguably the greatest hurdle relevant to feasibility is the will of the armed actor to include considerations of mental harm in decisions of attack. Currently, armed actors appear to demonstrate little political will to include incidental mental harm considerations, and it could be that the proffered culturally and contextually sensitive approach challenges political will even further due to its perceived complexity and the limited capacities of armed actors on the ground. As much of IHL does, implementation will rely on the armed actor’s willingness to interpret treaties in good faith.
Conclusion
The mental health toll of war cannot continue to be pushed aside, nor forced into unmalleable Western categorizations that obscure some lived realities. This article’s analysis of the principle of proportionality in attack and its implementation observes a weakness in the form of the continued hierarchization of physical harms. Yet, mental harms caused by war are just as prevalent and destructive as physical harms. Excluding excessive incidental mental harm from the principle of proportionality leaves civilian objects such as cars and animal livestock better protected than human mental health,Footnote 204 and risks hollowing out the principle and skewing its delicate balance between military necessity and humanity. This article does not call for legal reform; rather, it argues that a dynamic and culturally and contextually sensitive interpretation of the principle of proportionality, supported by further research, can fill the gap and stem the currently under-regulated infliction of excessive incidental mental harm in conflict.
This article is the first to invite reflections from the decolonizing global mental health agenda into the ongoing discussions about incidental mental harm and proportionality in attack. It has shown that efforts encouraging considerations of mental harm within decisions of attack in war must not be complicit in reinforcing a hegemonic universalization that privileges the experiences, theories and concepts of some at the expense of others. Rather, alternative knowledges and diverse conceptualizations of mental harm that hold relevance for the affected community themselves must be embraced. Knowledge of the socio-cultural deviations in how mental harm is experienced should be deepened and continually integrated into mainstream mental health dialogues – including those around the principle of proportionality in attack.
The present discussion provides a step towards addressing decolonial critiques of how incidental mental harm is engaged with for the purpose of the principle of proportionality in attack. The suggested culturally and contextually sensitive approach seeks to re-centre the lived experiences of affected communities in both law and practice. It hopes to encourage armed actors, policy-makers, researchers, and all those working in the conflict sphere to work towards the goal of integrating mental harm into the implementation of jus in bello proportionality in a way that is more coherent with a wider range of lived experiences of conflict and which will ultimately improve mental health protections in war.
Open access
