Introduction
In recent decades, States and military alliances have developed policies to better protect civilians in conflicts, and the operational importance of such approaches is increasingly accepted, notwithstanding recent pushback by the US government.Footnote 1 Simultaneously, it is known that gender affects individuals’ experiences of armed conflict.Footnote 2 A review of national, regional and international policies and doctrine indicates that gender-sensitive perspectives are increasingly incorporated into human security (HS) and protection of civilians (POC) doctrines.Footnote 3 These developments, however, often fail to go beyond the strategic and policy levels, lacking gender mainstreaming on the operational and tactical levels.
Especially in the planning and conduct of kinetic operations, the relevance of gender analysis is rarely included in relevant documents.Footnote 4 In the few instances where gender is mainstreamed into decision-making on the operational and tactical levels, such considerations are seen as operationally and strategically relevant – for example, to better protect a party’s own troops and to advance strategic goals.Footnote 5 In contrast, gender considerations are not perceived as relevant to discharging States’ obligations under international humanitarian law (IHL) in the conduct of hostilities, in particular during targeting.Footnote 6
The aim of this article is to illustrate that incorporating a nuanced understanding of gender-specific impacts of armed conflict is vital for States’ compliance with IHL. This can subsequently bridge the gap between high-level policies and the operational and tactical realities that are currently falling short of considering gender perspectives. A coherent incorporation of a gender lens into military doctrine can lead to a more institutionalized development of gender-sensitive rules of engagement (RoEs), standard operating procedures (SOPs) and operational planning more broadly. Crucially, it is not argued here that a gender-sensitive IHL interpretation is one of many potential interpretations of IHL; instead, it is shown that with today’s knowledge and capabilities, a gender-sensitive lens in military decision-making on all levels is indispensable to fulfilling IHL obligations. The focus thereby lies on the conduct of hostilities and obligations of conduct to prevent, limit and mitigate the effects of armed conflict on civilians.Footnote 7
The article starts by briefly summarizing recent policy and doctrinal developments which leave two gaps: (i) gender-specific considerations are only incorporated at the highest policy level and are not seen as legal obligations, and (ii) these policies, to a large extent, ignore or do not sufficiently develop how such considerations would inform operational and tactical decision-making in the conduct of hostilities, especially targeting. Then, by focusing on the core IHL principles of distinction, proportionality and precaution, it is shown that gender-blind decision-making is incompatible with good-faith implementations of IHL. The final section gives examples of how gender-sensitive considerations can enhance compliance with IHL.
The article does not provide a comprehensive overview of existing practice and relies on publicly accessible documents. Where relevant, examples of good practice or missed opportunities are used to illustrate the arguments made, with a focus on members and allies of the North Atlantic Treaty Organization (NATO).
Building on definitions from NATO and the United Nations, the term “gender” in this article refers to the social attributes associated with being male, female or of another gender, learned through socialization, that determine a person’s position in a given context.Footnote 8 Importantly, “gender” neither equates to “woman”, nor does it to “sex”, which is usually associated with physical or biological characteristics of bodies.Footnote 9 Gender analysis encompasses the experiences of all genders, including men and boys. This article primarily refers to the male/female gender binary, as it reflects the predominant framework currently incorporated in military doctrine, which largely overlooks the experiences of non-binary and gender non-conforming individuals. This is not meant to diminish the experiences of individuals in armed conflict identifying beyond the binary construct of gender.
This article is written with the goal of enhancing good-faith implementations of IHL. While the author is aware that bad-faith actors could abuse the recommendations made here in order to intentionally harm civilians, in particular in the context of data analysis, this will not be discussed as it is a general concern not exclusively related to gender-specific aspects.Footnote 10
Recent developments: Mainstreaming gender to win wars but not to enhance compliance with IHL
This year, the Women, Peace and Security (WPS) agenda turned 26. As part of the agenda, decision- and policy-makers have implemented measures to address the vulnerability of women and girls as victims of war, especially of sexual and gender-based violence (SGBV), and have called for increased female representation in armed forces and decision-making positions.Footnote 11 Over the course of a quarter of a century, however, the debate has rarely turned to the broader gendered effects of armed conflicts, and almost no attention has been paid to operational aspects and the conduct of hostilities, especially in the planning and execution of attacks.Footnote 12 The over-focus on SGBV has been criticized as it coincides with a lack of awareness of and disregard for other harms such as loss of livelihoods and homes, the spread of diseases, and displacement.Footnote 13
Some policy developments in recent years have succeeded in capturing broader gender-specific effects of conflict, moving beyond seeing women as vulnerable victims of SGBV. Many of these developments, with exceptions from Australia, NATO and the Nordic Centre for Gender in Military Operations (NCGM) in Sweden,Footnote 14 remain on the strategic level, lacking guidance on how such aspirations could be translated to the operational and tactical levels.Footnote 15 The UK Ministry of Defence (MoD), for example, states in its HS policy that access to health care must consider gender-specific services,Footnote 16 but while this is an important consideration, it is unclear how it can be operationalized when planning kinetic operations.
Arguments supporting the operationalization of gender often rely on the claim that such approaches advance a party’s own goals. In POC and HS policies as well as academic literature, gender-sensitive approaches are presented as a means towards achieving operational and strategic objectives, including “peace and security”, support from local civilian populations or the protection of a party’s own forces.Footnote 17 This can be situated in a broader trend of civilian harm mitigation policies serving a party’s own strategic aims.Footnote 18 Put differently, in the words of General McChrystal referring to the US campaign in Afghanistan, “We’re going to lose this fucking war if we don’t stop killing civilians.”Footnote 19 NATO, for example, notes that “[a]pplying a gender perspective and ensuring gender mainstreaming will be critical” to delivering its Military Instrument of Power.Footnote 20
While gender mainstreaming can indeed make operations more effective,Footnote 21 this argument is dangerous. It may lead to gender aspects only being taken into account when the benefits outweigh the costs, including for comprehensive training, new roles such as gender advisers (GENADs) or the development of new analysis tools. Additionally, commanders will be less convinced by the operational effectiveness of gender analyses in large-scale combat operations than in counter-insurgency operations which depend on the support of the local population. It is questionable whether and how gender mainstreaming can ever encapsulate a holistic approach, with the genuine goal of better protecting civilians, when it is only seen as a means towards achieving military success. A similar concern was raised in consultations for the global study on the implementation of the WPS agenda, where In the context of “women felt strongly that women’s rights should not be securitized” but should be considered as an end in themselves.Footnote 22
In addition to the general criticism of only incorporating gender considerations as a tool for strategic success, POC and HS policies often focus on harm that others cause to local civilians. These policies address SGBV or human trafficking committed by locals but do not, for example, consider the gendered secondary effects of conflict parties’ own kinetic operations, such as through the destruction of livelihoods or routes to water collection points.Footnote 23 Additionally, when gender considerations are mentioned, they remain very broad. For example, the NATO Allied Joint Doctrine stresses that a “gender lens should always be applied when analyzing the operating environment”, without clarifying what this entails in practice.Footnote 24 Such attempts to mainstream gender are not translated into the application of IHL: even though gender considerations are mentioned in NATO’s Joint Targeting Doctrine, they are not seen as relevant to complying with IHL, and while close cooperation between GENADs, legal advisers (LEGADs) and targeteers is deemed important, no reference to IHL is made in the context of gender considerations.Footnote 25 This is in line with gender being considered more relevant for civil-military operations where the aim is to build support, legitimacy and consent among the civilian population.Footnote 26
Gender considerations thus continue to be deemed relevant mainly in the context of peace operations, the rule of law and the rights of displaced persons, and when addressing human trafficking and SGBV.Footnote 27 If at all, reference is made to WPS resolutions or human rights law as legal sources, not IHL.Footnote 28 In its report Gender Makes Sense: A Way to Improve Your Mission, the NATO-accredited Civil-Military Cooperation Centre of Excellence only mentions human rights law, not IHL.Footnote 29 This is inversely mirrored by targeting doctrine and military manuals, which focus on the implementation of IHL, lacking language on incorporating a gender lens into the operational aspects of kinetic operations.Footnote 30 This blind spot makes it highly unlikely that gender is systematically mainstreamed into the operational and tactical decision-making levels and the development of RoEs or SOPs.
Hence, as noted above, recent positive developments aimed at better incorporating gender analysis largely leave two gaps: (i) gender-specific considerations are only incorporated into policies at the highest level and are not seen as legal obligations, and (ii) these policies, to a large extent, ignore or do not sufficiently develop how such considerations would inform operational and tactical decision-making in the conduct of hostilities, especially targeting.
Core IHL principles and the gendered effects of conflict
The main legal framework that regulates the conduct of hostilities and protects all victims of armed conflict was developed in the nineteenth and twentieth centuries, when the gender-specific effects of armed conflict were not widely understood and there was almost no representation of women or individuals not identifying under the male/female binary in IHL legal drafting and decision-making.Footnote 31 Unsurprisingly, relevant treaty laws are largely silent on gender, except when they aim to protect women’s “honour” and their role as mothers. These provisions have been criticized as they mainly see women in relation to men and fail to capture how gender affects individuals’ experience of conflict more broadly.Footnote 32 The term “gender” is absent from IHL treaty provisions.Footnote 33 While IHL provisions on the conduct of hostilities are gender-blind and seem to provide everyone with the same protections regardless of gender, the framework has been described as “a deeply gendered body of law”Footnote 34 as it cements pre-existing inequalities. More generally, feminist writers have argued that international law’s purportedly neutral standards tend to assume a male perspective.Footnote 35
However, ignorance with respect to the gendered impacts of conflict at the time of the drafting of IHL treaties does not mean that warring parties today, with their current knowledge and capabilities, do not need to incorporate measures to account for these impacts. As an area of public international law, IHL’s treaty texts are to be interpreted “in good faith in accordance with the ordinary meaning to be given to the terms of the treat[ies] in their context and in the light of [each treaty’s] object and purpose”.Footnote 36 The object and purpose of IHL is to balance military necessity with humanity, thereby limiting the humanitarian consequences of armed conflicts by restricting the means and methods of warfare.Footnote 37 For example, if a warring party has the means and capabilities to use a high-precision rocket which would avoid incidental civilian harm in a given attack, it could not, without a good reason, resort to less precise weapons that would inevitably cause civilian casualties only because such more precise weapons were not available in 1977.Footnote 38 Additionally, IHL evolves and develops through interpretation and practice, such as by adapting to technical advancements and new realities on the battlefield. Further, because most IHL treaty rules only apply in international armed conflicts, the development of customary international law is particularly important in IHL to also regulate non-international armed conflicts.Footnote 39
This section demonstrates that implementing IHL while disregarding the role of gender, on the grounds that IHL treaty law is largely gender-blind, is difficult to reconcile with a good-faith interpretation of IHL’s object and purpose.
In February 2010, a group of what US soldiers perceived to be military-aged males (MAMs) drove in three vehicles through a Taliban stronghold in Afghanistan. A US drone strike targeted the convoy, killing and wounding its occupants, which in fact included women and children. Investigations later revealed that the group was made up of civilians and evidence indicating that the convoy was not hostile was “ignored or downplayed”.Footnote 40 The investigations, however, did not sufficiently address the role that gender bias played in the misidentification of the targets; to the contrary, in one investigation, a US Army general blamed the failure to spot women or toddlers as a main mistake. The communication from the Predator drone’s crew “led [the US troops in the area] to believe that the vehicles contained only armed [MAMs]”.Footnote 41 This is problematic because even if all passengers had been MAMs, this would not have stripped them off their protection as civilians.
The 2010 incident shows that even when there is no official policy to target MAMs based on their gender, unaddressed bias can lead to similar effects. Decision-makers falsely interpreted any behaviour or evidence in light of the potential targets being MAMs. Evidence used to positively identify the targets consisted of “the weapons we’ve identified and the demographics of the individuals”, despite there only being unconfirmed suspicions of a rifle.Footnote 42 In fact, the civilians did not carry any weapons, and while decision-makers interpreted the fact that multiple vehicles were driving together as suspicious behaviour, this was in fact being done so that the civilians could support each other in case one vehicle broke down. This example is not an isolated case – according to US investigations, misidentification was the primary cause of civilian casualties in Afghanistan.Footnote 43
Distinction
Under the principle of distinction, warring parties must at all times distinguish between combatantsFootnote 44 and civilians. Attacks may only be directed against the former or civilians who directly participate in hostilities.Footnote 45 As with all core IHL principles, the relevant provisions codifying the principle of distinction (Articles 48 and 51(2) of Additional Protocol I (AP I)) do not mention gender and are hence in theory gender-neutral: no matter their gender, no civilian must be targeted unless and for such time they directly participate in hostilities.
In most conflicts, there are more direct male than female casualties.Footnote 46 This may be because more men are combatants, directly participate in hostilities or frequent the vicinity of military objectives – unequal casualty numbers do not directly indicate discrimination or bias in targeting practices. The example mentioned above, however, shows how gender bias can lead decision-makers to interpret patterns of life and potential evidence in light of, in this case, maleness: in the 2010 drone strike, evidence was (mis-)interpreted to confirm “hostile intent”.Footnote 47 An officer involved described the unchallenged prejudices that impacted subsequent reasoning: “We all had it in our head, ‘Hey, why do you have 20 [MAMs] at 5 a.m. collecting each other?’… There can be only one reason and that’s because we’ve put [US troops] in the area”Footnote 48 (referring to US Special Forces being in the vicinity). Through bias, it was more readily assumed that MAMs had lost their protection against attack, even though they did not meet the threshold of direct participation in hostilities (DPH).Footnote 49
This undermines the principle of distinction: for individuals who are not members of States’ armed forces, civilian status must be presumed unless proven otherwise, irrespective of gender.Footnote 50 Civilians are protected from being the object of an attack unless and as long as they directly participate in hostilities.Footnote 51 If gender (maleness) were to become a main identifier to conclude DPH, this process would be turned on its head: MAMs would be presumed to be directly participating in hostilities, requiring evidence to disprove it. Taking sex as a crucial identity signifier, especially for targeting, has been described as a form of gender-based violence.Footnote 52 The bias of naturally perceiving MAMs as a threat is particularly dangerous in the context of signature strikes, where targeted individuals are not identified but targeting is instead based on pre-identified signatures of behaviour that are believed to indicate hostile intent or activity.Footnote 53
New Zealand’s Manual of Armed Forces Law notes that commanders should consider gender and age in the targeting process. While the manual concedes that “it is not allowable to target every [MAM] on that basis alone”, old women are “unlikely” to directly participate in hostilities.Footnote 54 To determine DPH, decision-makers should consider “the activity, weapons (if visible), clothing, gender and age of the individual”.Footnote 55 While such guidance does not call for the revocation of civilian status for all MAMs, it can encourage, strengthen and justify pre-existing biases that lead to the above-explained misidentification in targeting because decision-makers may be less diligent when concluding that an MAM has lost their protection against direct attack.
Even though, according to a study from 2019, women play some leadership role in at least 28% of the armed groups surveyed in the Women in Armed Rebellion Dataset,Footnote 56 they continue to be often perceived as inherently non-violent, peaceful civilians and vulnerable victims by virtue of their gender.Footnote 57 This can lead to a presumed overlap of civilian status and femaleness.Footnote 58
Proportionality
According to the principle of proportionality, when conducting an attack, the expected incidental loss of civilian life, injury to civilians, damage to civilian objects or a combination thereof must not be excessive in relation to the concrete and direct military advantage anticipated.Footnote 59 Similar to the principle of distinction, the relevant provisions codifying the principle of proportionality (Articles 51(5)(b) and 57 of AP I) do not expressis verbis require parties to a conflict to consider the gendered effects of attacks. It will be shown in this section, however, that a failure to do so impedes compliance with this principle. This is because proportionality involves subjective assessments on the part of the “reasonable commander”Footnote 60 with regard to (i) what is considered as incidental civilian harm, (ii) which harms are considered to be foreseeable and (iii) when expected harm is considered to be excessive in relation to the anticipated military advantage. Gender-blind approaches lead to an underemphasis of harms that are disproportionately experienced by women, and this is amplified by the fact that the proportionality principle only considers effects from specific attacks, not the military campaign as a whole.
What types of harm are considered?
Individuals experience armed conflicts differently, and general trends can be observed with regard to gender.Footnote 61 While mortality in conflicts is difficult to assess, men are overall more likely to be directly killed by bombing, shelling or explosive weapons,Footnote 62 including landmines.Footnote 63 A similar pattern has been observed for direct casualties from chemical weapons attacks in Syria.Footnote 64 Casualties from cluster munition remnants are likewise higher among men than women, though the highest casualty number by far is children.Footnote 65
In contrast to direct casualties, women are on average impacted more by subsequent effects from exposure to weapons. Once injured, women and girls are more likely to later die than men and boys and are in general more affected by limited access to health care.Footnote 66 This may indicate that women face barriers to timely and adequate treatment. In Afghanistan, health services were severely restricted for women, impacting their treatment from exposure to explosive remnants.Footnote 67 Women survivors from Hiroshima and Nagasaki faced almost double the risk of developing and dying from cancer compared to men.Footnote 68 Use of Agent Orange in Vietnam resulted in high levels of miscarriages and premature births, and generally impacted women’s reproductive health.Footnote 69
Women additionally face risks during displacement such as human trafficking or SGBV, impeded access to reproductive health care, economic insecurities, domestic violence or insufficient access to resources, leading to higher risks of malnutrition. While everyone in conflict-affected regions is at risk of these harms, women are disproportionately affected.Footnote 70 Data suggests that in humanitarian settings, women are less likely to be formally employed or in paid work: in “fragile and conflict-affected” countries, 66% of men are in paid work as opposed to 44% of women, and in countries with protracted conflict, the latter percentage is below 20%.Footnote 71 Depending on the conflict, women can be more confined to their homes and more dependent on their own livestock or crops, with fewer opportunities for finding employment.Footnote 72 Women are therefore more exposed to food and financial insecurity, poor sanitation and long-term health impacts such as post-traumatic stress disorder and disability, making it more difficult for them to cope with property damage, diseases or injuries.Footnote 73 While women are more affected by such indirect and reverberating effects than men,Footnote 74 it is also observed that overall more people (of all genders) die from such indirect effects than from the direct effects of war.Footnote 75
In the context of genocide, it has been observed that “slow deaths” such as deaths resulting from sustained sexual and physical abuse, poor living conditions, suicide following mistreatment or the infection of victims with HIV are considered less important than “fast deaths” such as direct killings. With women mainly facing “slow deaths”, Sareta Ashraph argues that women are “denied equal protection before the law”, maintaining that a failure to equally recognize “slow deaths” is discriminatory.Footnote 76 The principle of proportionality is substantially different from the prohibition against genocide as civilian harm in proportionality assessments is incidental and therefore not intended; still, it can be similarly critiqued that it is discriminatory to systematically disregard harms that disproportionately affect women.
Many of these indirect harms do not fall within the three categories mentioned in the treaty-law provision of proportionality: casualties, property damage and injuries. Still, some States adopt broader approaches, using terms such as “suffering and destruction”,Footnote 77 “adverse effect”Footnote 78 or “possible harmful effects”.Footnote 79 These concepts are theoretically capable of capturing harms beyond direct casualties, property damage and injury. The United States, however, in addition to excluding indirect economic harm,Footnote 80 has stated that “mere inconveniences” do not need to be considered in proportionality assessments; such inconveniences include “inevitable scarcities of foodstuffs and services”.Footnote 81 Importantly, these inconveniences – which can be life-threatening – affect women more than men due to the former’s often lower societal and economic status. While mental harm is capable of falling under injury,Footnote 82 its inclusion in proportionality assessments when unrelated to physical injury has continuously been challenged for reasons including the difficulty of estimating it.Footnote 83
Without disregarding ongoing debates regarding the scope of incidental harm under the principle of proportionality, as of yet States have to a large extent not adopted measures to systematically consider harms beyond death, physical injury and property damage in proportionality assessments.Footnote 84 A major difficulty cited in this regard is the foreseeability of indirect harms.
Reasonably foreseeable harm
Only those harms, including indirect or reverberating effects, that are reasonably foreseeable at the time of attack need to be considered by the reasonable commander.Footnote 85 Incidental civilian harm caused by military action is considered foreseeable where it is “based on information reasonably available from all sources in the circumstances”.Footnote 86 This is contingent on what data has been collected or what external data has been considered to be relevant, including context-specific data from humanitarian actors and general gender trends in the operating region. The weight that the commander places on different types of data also depends on relevant doctrine and policy as well as the commander’s training in and awareness of gendered harms. Hence, what is considered foreseeable is contingent on training, intelligence-gathering and willingness to engage with known risks for the civilian population, including their gendered dimensions. In practice, there currently persists what the International Committee of the Red Cross (ICRC) calls an “invisibility barrier”: while it is known – and foreseeable – that gender inequality persists and often worsens during conflict, there is a lack of data collection and analysis of context-specific gendered impacts.Footnote 87
The question of what can be expected from a reasonable commander depends on how reasonableness is assessed. It has been shown that culture influences what is considered reasonable for a commander in a given context, and therefore in proportionality assessments.Footnote 88 Previous criticisms of the “masculinization” of the conduct of hostilities rulesFootnote 89 and broader observations that diverse perspectives, especially those of women, were lacking in the development of IHL, and to some extent are still lacking in its application and implementation today,Footnote 90 suggest that the “reasonableness” that is expected from a commander when making proportionality assessments comes from a perspective that cements the disregard of indirect harms which disproportionately affect women and girls. Therefore, a broader shift in accepting that the gendered dimension of incidental harm is foreseeable will also impact, in the long run, what a commander is reasonably expected to take into account when evaluating the expected incidental civilian harm of an attack.
Excessiveness
An attack is disproportionate when the expected incidental civilian harm is excessive in relation to the anticipated military advantage. There is no international guidance on when excessiveness is reached; hence, the way in which incidental civilian harm is balanced against military advantage reflects the subjective priorities of decision-makers, including those involved in designing RoEs and SOPs or the commander deciding a specific attack.Footnote 91
When only direct casualties are considered to be severe incidental harm or when a gender-blind approach is taken with regard to damage to civilian infrastructure, this leads to a lack of consideration of harms that particularly affect women. Consider, for example, a scenario in which the expected incidental damage to a hospital from an attack is quantified at 5%, as only one part of the building will be damaged.Footnote 92 With other functioning hospitals in the region, the expected harm (and the reverberating effects stemming from the 5% decline in services) is seen as not excessive when weighed against the anticipated military advantage. However, what if that 5% of the hospital houses the only obstetric care unit in the whole region? In such a case, women would be disproportionately affected by the attack.Footnote 93
To conclude, whether an attack is deemed proportionate depends on (i) which harms are considered to be not only “mere inconveniences” but relevant under the principle of proportionality, (ii) which harms are considered to be foreseeable and are hence taken into account, and (iii) the severity attributed to harms. All three factors are interrelated: which harms are taken into account influences when the excessiveness threshold is met. Consider a situation in which someone with a continuous combat function is killed through an air strike on their house, destroying the house while their civilian spouse and child are not home. Depending on whether the house was used for military purposes, the property damage could factor into the proportionality assessment. However, the risks that the family face because of displacement resulting from the destroyed home, economic hardship potentially leading to disruption of the child’s education, or mental harm will most likely not be considered.Footnote 94 Hence, even indirect harms that are attributable to one specific attack are often considered too remote by decision-makers to play a role in proportionality assessments. Another example would be heightened risk of violence, including SGBV, when water or wood collection routes change after an attack causes infrastructure damage.
Therefore, harms disproportionately faced by women are currently systematically less considered in IHL proportionality assessments because indirect harms are considered to be more difficult to foresee and less severe, and because the proportionality treaty rule only mentions deaths, injury and property damage. This coincides with and is amplified through the fact that the principle of proportionality has a limited scope because it only applies to specific attacks, not the overall consequences of conflict.Footnote 95
While cumulative harms and their gendered dimensions are to some extent foreseeable as general trends in armed conflicts,Footnote 96 warring parties will most likely not consider them in their proportionality assessments for specific attacks as such harms will be seen as too remote. Therefore, IHL has a blind spot with regard to the indirect, accumulating and long-term harms caused by conflicts – i.e., harms that disproportionately affect women. Even under a broad understanding of reverberating effects, many harms would still fall outside the principle of proportionality as they often cannot be attributed to specific attacks. This includes starvation following economic hardship, diseases that spread because of general poor hygienic and humanitarian conditions, or protracted displacement.Footnote 97 This is why, moving forward, an approach is warranted that takes cumulative civilian harm into account on the strategic level; such an approach is currently being developed.Footnote 98
Precaution
In addition to complying with the principles of distinction and proportionality, a warring party must take all feasible precautions to avoid, and in any event minimize, incidental civilian harm.Footnote 99 Feasible precautions are those that are considered practicable or practically possible while considering the specific circumstances at the time. A commander must “obtain the best possible intelligence”, which may include information about important civilian infrastructure (specifically protected objects), the concentration of civilians or the natural environment.Footnote 100
Not considering gendered trends when issuing warnings to the civilian population may put women at a higher risk. For example, a gendered digital divide has been observed in some regions, and impeded access to communications technologies and the internet might create barriers to warnings reaching certain groups.Footnote 101
Patterns of gendered incidental civilian harm in conflicts are known. Gender inequalities persist worldwide, with conflicts usually seeing higher rates of inequalities. For example, as women often take on caring roles in families and face heightened risks during movement, this leads to less mobility and can impact their ability to flee from attacks.Footnote 102
US doctrine includes reference to gender perspectives enabling “commanders to examine risks and trade-offs to determine [courses of action] with fewer unintended consequences to the [host nation’s] population”.Footnote 103 Hence, if taking such perspectives into account is a way to cause “fewer unintended consequences”, this is a measure that is practicable or practically feasible and therefore should be incorporated into the precautionary measures taken. The US Center for Army Lessons Learned has noted that the term “MAM” might unintentionally suggest hostile intent as it implies that “the individuals [in question] are armed forces and therefore legitimate targets”.Footnote 104 However, it appears that this has not provoked substantive changes to the targeting procedure or the implementation of safeguards aimed at mitigating gender bias. Even though General McChrystal reportedly banned using the term “MAM” after the 2010 strike referred to above, it continued to be used in official military documents.Footnote 105 Having knowledge of the deadly effects of gender bias in the targeting process but not addressing them seems difficult to reconcile with the requirement to take all feasible precautions. Failing to address biases additionally affects post-strike battle damage assessments (BDAs) as civilian MAMs might not be counted as such. This in turn may affect potential lessons learned and the identification of patterns of civilian harm.
Where commanders are aware or should have been aware of mistakes made, such as the misidentification of MAMs as legitimate targets, this may constitute constructive knowledge of such shortcomings. Failure to address these issues may violate the principle of taking all feasible precautions to avoid or minimize incidental civilian harm. When mistakes happen repeatedly or there appears to be a pattern of mistakes, the reasonable commander should assume that such mistakes may happen again and investigations must be commenced.Footnote 106 Therefore, gender-specific patterns of harm or mistakes that result from biases must be addressed accordingly.
Gender-sensitive approaches to fully implementing IHL
It has been shown that gender influences individuals’ experiences of conflict, including the risks they face. Therefore, it is crucial for warring parties to incorporate known trends or context-specific intelligence into their decision-making in order to fulfil their IHL obligations. This includes the principle of precaution, in turn affecting the principles of distinction (e.g., misidentification) and proportionality (e.g., not considering certain harms because they are deemed not foreseeable). When information about harms particularly affecting women is not collected or is disregarded, this affects the principle of proportionality as such harms will not be incorporated into the proportionality assessment.
Despite this, most law of war manuals lack exactly such an analysis.Footnote 107 In recent years, measures have been developed that have the potential to help States incorporate gender considerations not only on the strategic but also on the operational and tactical levels. This includes gender analyses to better understand the civilian environment, including data disaggregated by gender.Footnote 108 However, most States have as of yet failed to develop robust plans that are sufficiently comprehensive to provide guidance for the operational or tactical levels, falling short of fulfilling IHL obligations in this regard.
To address the currently existing blind spot, measures could be implemented to mitigate bias in the targeting process. Because of the bias to derive DPH from maleness, it is insufficient to ban terms such as “MAM” or to implement policies that simply confirm that not all MAMs are legitimate targets. Instead, targeting checklistsFootnote 109 could include a requirement for decision-makers to apply greater scrutiny when analyzing evidence (e.g., presence of weapons) that supposedly identifies MAMs as legitimate targets. Furthermore, all staff involved in targeting processes should receive training on biases, including in pattern-of-life assessments.
Additionally, data from BDAs and other sources disaggregated by gender could help to identify patterns of harm and subsequently contribute to lessons learned.Footnote 110 Data on civilian infrastructure and access to it, including gendered dimensions such as access to obstetric care, needs to be included in the development of no-strike or restricted strike lists.Footnote 111 Where, for example, a party to the conflict has information on the accessibility of hospitals, it must be ensured that attacks do not render the civilian population’s access to essential care impossible. For instance, if two hospitals serve a neighbourhood but only one provides obstetric services, it is necessary to ensure not only that the latter facility is protected from destruction, but also that civilians can continue to reach it.
To address the harms that especially women in Afghanistan faced when their homes were destroyed, Jody Prescott has suggested prohibiting air strikes, other than in self-defence, against home compounds.Footnote 112 In the context of environmental harm, the ICRC has recommended context-specific risk analyses that also consider how the effects vary depending on individuals’ characteristics, including gender.Footnote 113 Such an approach would be equally important for armed conflicts more broadly, not only with regard to the environmental dimension.
Additionally, different armed forces have introduced GENADs on the operational level and gender focal points on the tactical level.Footnote 114 Some of these are involved in target development and/or directly report to the commander,Footnote 115 or are part of the command group, together with the commander, chief of staff and LEGAD, among others.Footnote 116 Some doctrine suggests that the main role of GENADs is to raise awareness on the situation of women and children as civilians,Footnote 117 but this is insufficient – gender is not a synonym for “women and children”, and relevant considerations need to go beyond protection from SGBV. This includes considering the heightened risks that men and boys face, such as through misidentification.Footnote 118
Furthermore, GENADs occupy roles distinct from LEGADs as they are not involved in legal assessments during targeting.Footnote 119 This is problematic because, as has been shown above, gender-blind approaches can lead to a failure to comply with IHL. GENADs should be seen as essential for compliance with IHL and should therefore participate in targeting decisions at the highest level, as well as dynamic targeting, and should be involved in developing SOPs and RoEs.Footnote 120 Conflict parties should consider having specialized LEGADs with additional expertise in gender, or alternatively GENADs with sufficient IHL expertise. This would mean that LEGADs with gender expertise would be involved in the training of armed forces and exercises; they would advise on the legality of attacks, which would include gender-specific aspects such as those discussed above in all steps of the decision-making process, including for proportionality assessments and when taking all feasible precautions to avoid or minimize incidental civilian harm. In practice, this would mean that such experts would advise on RoEs, SOPs, target packages, and no-strike and restricted target lists, as well as on specific targeting decisions in the moment. Only by incorporating gender expertise into all of these steps would the mainstreaming of gender considerations on the operational and tactical levels be possible. Conversely, if GENADs were only involved in discussions at the strategic level and then in the assessment of individual targeting decisions, it is unlikely that a force would be able to comprehensively incorporate gender analyses into its decision-making.
Having LEGADs with gender expertise, or GENADs with sufficient IHL expertise, present during decision-making and when evaluating the effects of attacks, including through BDAs, makes it more likely that these advisers will identify specific gendered harms that should be addressed. For example, when MAMs are misidentified as potential targets, experts will be more likely to spot such patterns and will be able to advise on appropriate measures to address existing biases in the targeting process, including questions of bias in targeting checklists and training. Such an approach of involving LEGADs with gender expertise from early on, including in training exercises, will help mainstream considerations of the gendered harms of kinetic operations and will make it more likely that such processes will function well when implemented in operations.
In all of these considerations, the argument that it is not feasible to mainstream gender on the operational and tactical levels because insufficient data is available is weak. While States’ obligations depend on the foreseeability of harm and hence potentially on the available data, general patterns of gendered harm are already known and data on such patterns of harm, the demographics of the affected civilian population and context-specific issues are often available from humanitarian and international organizations.Footnote 121 For example, when access points for spring water or basic goods are destroyed, civilians have to take a different or longer route to collect these goods, and this puts women and children at greater risk.Footnote 122 Therefore, the claim that gendered effects are unknown is unconvincing if it is solely based on warring parties’ failure to collect their own intelligence. Sufficient information is available such that “certain reasonable operational assumptions” can be made regarding the effect of gender inequality on individuals’ experiences of conflict and the resulting risks.Footnote 123
Additionally, States could invest in and develop tools for gender analysis that in the long term will allow them to better understand and mitigate harm. Such tools could, for example, track direct and indirect casualties and other data from BDAs disaggregated by gender and type of weapon. They could also analyze property damage and displacement dynamics and incorporate data from humanitarian organizations on the spread of diseases and civilians’ access to essential resources, disaggregated by gender. This would allow decision-makers to better asses the civilian environment, and for example take access routes to essential services such as health care and basic resources into account when developing and reviewing target packages and no-strike and restricted target lists, and when planning specific attacks.
In addition to a periodic review of RoEs, SOPs and planning documents, reviews should be triggered when patterns of gendered harms are observed. This data should also inform feedback loops used to develop lessons learned for ongoing and future operations, training and the foreseeability of incidental civilian harm.
The exact measures that need to be taken will depend on the capacity and structure of the armed forces or non-State armed group implementing the changes, the specific operation and the information available. The Geneva Centre for Security Sector Governance (DCAF) and the Center for Civilians in Conflict (CIVIC) have developed a gender-responsive civilian harm mitigation checklist for armed forces that builds on recommendations to fully integrate gender-sensitive considerations at all stages of their actions. These range from integrating gender-responsive civilian harm mitigation into all military policies, doctrines and mission objectives to recommendations on training, data collection and analysis, community engagement, and investigation and accountability systems.Footnote 124
Conclusion
Policies developed over recent years indicate States’ willingness to better understand and mitigate gender-specific harms in armed conflicts, for example by conducting gender analyses to enhance the understanding of the civilian environment. Most of these policies, however, remain at the highest strategic level, falling short of acknowledging the role of gender considerations in fulfilling IHL obligations; instead, such considerations are mainly seen as a policy measure for fighting more successful wars. The present article has shown how this blind spot can hamper compliance with IHL, including with the core principles of distinction, proportionality and precaution.
Today, general trends with regard to the gender-specific impacts of conflict are known and available and could therefore be used by parties to a conflict to mitigate incidental civilian harm. Therefore, gender is not merely one of many lenses through which IHL can be interpreted; instead, it has been shown that with today’s knowledge and capabilities, gender considerations in military decision-making on all levels are indispensable to fulfilling IHL obligations. The acknowledgment that gender-sensitive approaches are critical to complying with IHL in the planning of kinetic operations will help to bridge the existing gap between high-level policies and practice on the ground. This is because it will be necessary for States to implement gender considerations into their law of war manuals, and subsequently into SOPs, RoEs and decision-making on the operational and tactical levels.
Future research should further investigate the intersection of gender with other characteristics that influence individuals’ experiences of armed conflict, including age, socio-economic background, sexuality and race. Additionally, analyses and recommendations must recognize the experiences of all individuals in armed conflicts, including individuals not identifying under the binary construct of male/female. Furthermore, more specific recommendations need to be developed for issues such as urban warfare and different types of weapons, including new technologies and artificial intelligence, which risk amplifying already existing biases in the targeting process.
Open access
