Introduction
Over twenty-five years ago, United Nations (UN) Security Council Resolution 1325 on “Women, Peace and Security” called upon the international community to take comprehensive steps to increase the protection of women and girls from the impacts of armed conflict. The resolution specifically noted “the need to implement fully international humanitarian and human rights law that protects the rights of women and girls during and after conflict”.Footnote 1 Full implementation of international humanitarian law (IHL) includes the special protections afforded to women under the 1949 Geneva ConventionsFootnote 2 and Additional Protocol I (AP I)Footnote 3 to those treaties.Footnote 4
Among these special protections, Article 76 of AP I, on “Protection of Women”, requires that women be accorded “special respect” and that they be protected in particular against rape, enforced prostitution and indecent assault.Footnote 5 Although neither the provisions of AP I nor the influential 1987 Commentary on the Additional Protocols (AP Commentary)Footnote 6 by the International Committee of the Red Cross (ICRC) establish an explicit link between this special respect and the protections for the civilian population and individual civilians set out in Articles 51 (“Protection of the Civilian Population”) and 57 (“Precautions in Attack”) of AP I, this article argues that it should be applicable to those protections.
To determine whether such an interpretation would be appropriate, this article first assesses Article 76 of AP I under the standards for treaty interpretation set out in the Vienna Convention on the Law of Treaties (VCLT),Footnote 7 looking to the plain language of Article 76Footnote 8 and instances of State practice.Footnote 9 It is worthwhile to assess instances of State practice in detail, because this work provides insights into some of the challenges that armed forces face in effectively operationalizing understandings of the gendered impacts of armed conflict. The article then briefly examines Article 76’s negotiation history to confirm that such an interpretation was not dismissed by the different groups of experts negotiating the language of the provision.Footnote 10
Next, the article turns to Articles 51 and 57 of AP I, and explains how the scope of these provisions’ coverage of both civilian populations in general and individual civilians provides space for States Parties to take steps to implement the special respect set out for women in Article 76.Footnote 11 This is particularly so in the context of their obligations to avoid, and in any event minimize, the negative impact of military operations and to take precautions in attack, and in applying the principle of proportionality, including from the perspective of reverberating effects when substantiated by gender-inclusive intelligence. In conclusion, this article argues that Article 76(1)’s special respect for women requires consideration of gendered harms to comply with Articles 51 and 57, and that for this to be practicable, there must be significant expansion in the collection and analysis of gender-disaggregated data by military units in their areas of operation than exists today.Footnote 12
Article 76, protection of women, and analysis under the Vienna Convention on the Law of Treaties
The literature interpreting and assessing Article 76(1) since the time AP I came into effect shows some evolution in its understanding. The AP Commentary notes that the original ICRC draft of what became the first paragraph of Article 76 was only changed slightly during the negotiations at the Diplomatic Conference.Footnote 13 It further explains that the purpose of this article was to expand the coverage of the protections in Article 27 of Geneva Convention IV (GC IV) to all women impacted by armed conflict, and to incorporate developments in international human rights law that prohibited convicted pregnant women from being executed.Footnote 14 Article 27’s specific protections for women are narrower than those eventually made part of AP I Article 76, since Article 27 provides only that “[w]omen shall be especially protected against any attack on their honour, in particular against rape, enforced prostitution, or any form of indecent assault”.Footnote 15
In the almost forty years that have passed since the AP Commentary was published, writers have taken a variety of approaches in addressing the meaning and application of Article 76(1). One common approach shown in the literature is to assess this provision as favourable for women, and as an important part of an overall strengthening of the protections afforded to them.Footnote 16 Another common approach, however, is critical of the provision as reflecting outmoded and discriminatory ways of thinking about women’s gendered roles in society, such that the gravamen of sexual violence against women is mischaracterized as an offence against honour rather than against the person.Footnote 17 Some writers are concerned that the focus on sexual violence suffered by women is not complemented by a similar focus on sexual violence suffered by men.Footnote 18
These treatments, whilst thoughtful and cogent, have one thing in common: they largely do not address whether this special respect is only applicable in strengthening protections for women within the contours of Article 76, or whether it might be reasonably interpreted to apply to other provisions of AP I in which it is not explicitly mentioned, such as Articles 51 and 57. However, writing two years before the AP Commentary was published, Krill’s article on IHL’s protection of women is an exception to this trend.
In her article, Krill surveyed the many areas of the Geneva Conventions and AP I in which the formal equality of men and women is complemented by differentiated protections for women that promote their equal treatment, including AP I Article 76.Footnote 19 She noted that
the prohibition of discrimination is not a prohibition of differentiation. It is for this reason that distinctions are prohibited only to the extent that they are unfavourable. Equality could easily be transformed into injustice if it were to be applied to situations which are inherently unequal and without taking into account circumstances relating to the state of health, the age and the sex of protected persons.Footnote 20
Krill argued that the principle of differentiated treatment is not limited to only those instances in which differentiated protections are specified in IHL. As she saw it, “[a]n express reference tends to strengthen the scope of the principle, rather than to limit is application, and differentiated treatment is accorded to women even if it is not explicitly mentioned”.Footnote 21
The VCLT provides an objective and widely accepted methodology to test Krill’s argument. Under Article 31 of the Convention, an interpretation must be in good faith and consistent with the ordinary meaning of the treaty’s terms, in the proper context of the treaty, and consistent with the object and purpose of the treaty. The context includes all the treaty text and certain agreements made between States Parties to the treaty related to its conclusion, as well as any subsequent agreements between States Parties related to the treaty’s interpretation or application, and any practice of States Parties which would show agreement between them regarding its interpretation.Footnote 22 If this first step of the analysis does not sufficiently resolve ambiguity regarding the proper interpretation of the treaty, then under Article 32, it is then necessary to look to the negotiation history of the treaty.Footnote 23
VCLT Article 31: General rule of interpretation – plain language
To determine whether the special respect to be accorded to women in AP I Article 76(1) is limited to protection from the instances of sexual violence listed within the provision and the protections for pregnant women and mothers of small children under Articles 76(2) and 76(3), it is helpful to compare the text of Article 76(1) with the text of GC IV Article 27, on which the former was in part based. As previously noted, GC IV Article 27 provides that “[w]omen shall be especially protected against any attack on their honour, in particular against rape, enforced prostitution, or any form of indecent assault”.Footnote 24 Consistent with this linkage between heightened protection and sexual violence, the 1958 Commentary on GC IV (which has since been updated) describes “attacks on women’s honour” as rape and other forms of sexual assault, and sees the prohibition against these attacks as safeguarding the respect for family life, centred on “marriage ties and that community of parents and children which constitutes a family”.Footnote 25
The text of AP I Article 76(1) does not simply restate this formulation. Importantly, although Article 76 requires that women be particularly protected from sexual violence, it does not say that the required “special respect” is to be accorded to them only because they are at risk of sexual assault.Footnote 26 The two are separate, joined by a conjunction; “special respect” is not qualified by the offences against which women are to be particularly protected. Finally, looking at Article 76 as a whole, its subsequent paragraphs enumerate specific protections given to women consistent with this “special respect,” but these do not limit the breadth of that term.Footnote 27 Thus, “special respect” could be reasonably interpreted as applying to other AP I provisions related to the protection of women. The next step under VCLT Article 31 is to see whether any States Parties have done so in the context of AP I Articles 51 and 57, either explicitly or as part of the protections afforded civilians in general or as individuals, or implicitly in effect.
VCLT Article 31: General rule of interpretation – State practice
The spectrum of official acts of a country that can be considered State practice is quite broad.Footnote 28 For evidence of State practice in applying AP I Article 76(1)’s special respect for women in armed conflict, two types of documentary material are particularly germane: national IHL manuals and doctrine applicable to the uses of military force in instances where AP I Articles 51 and 57 are relevant.
National IHL manuals are recognized as fundamental references for military organizations on IHL.Footnote 29 In his thorough survey of the development of national IHL manuals, Garraway recognized that they have become important evidence of State practice.Footnote 30 He also cautioned, however, that although they could
indeed be used as part of the lawmaking process … their evidential value should not be overstated. Their purpose is ultimately practical ‒ to assist the soldiers, sailors and [airmen] on the frontline to “temper the harshness and cruelty of combat” by abiding by the law.Footnote 31
As to doctrine, there is no universal definition, but that used by the Australian Defence Force (ADF) is both succinct and substantive: “Military doctrine describes fundamental principles that guide actions by armed forces to achieve their objectives. While authoritative, doctrine requires judgement in application.”Footnote 32 Doctrine serves other important purposes as well, such as providing a basis for educational curricula and training evaluation and allowing militaries to prioritize limited funding for the most important mission sets.Footnote 33 The Nordic Centre for Gender in Military Operations (NCGM), the Swedish Red Cross and the ICRC have recognized the importance of including a gender perspective in military doctrine, including military IHL manuals, in order to ensure more complete institutionalization of gender considerations across the range of military activities.Footnote 34
National IHL manuals
Turning to State practice, a sampling of IHL manuals of certain AP I States Parties with fairly recent combat experiences shows little explicit development beyond restating AP I Article 76(1)’s requirement of women being the object of special respect. The UK Joint Service Manual of the Law of Armed Conflict (UK Manual), for example, quotes and cites Article 76,Footnote 35 and restates the outdated GC IV Article 27 formulation of women being protected “against any attack on their honour”.Footnote 36 “Special respect” is not further defined. The UK Manual does mention women’s different physical and medical requirements, noting, for example, that “[d]ue regard must be had to women’s physical strength … and the special demands of biological factors such as menstruation, pregnancy, and childbirth”.Footnote 37 However, other discussions of the treatment to be accorded to women largely restate the rules already found in the Geneva Conventions and their Additional Protocols.Footnote 38
The French Manual of the Law of Military Operations (French Manual) notes that women “enjoy special protection”, citing Article 76Footnote 39 and France’s joint efforts with IHL and human rights groups to develop training on strengthening women’s protections during conflicts and increasing women’s role in peacemaking processes.Footnote 40 It does not elaborate further on France’s views regarding the language of Article 76. The French Manual also notes Security Council Resolution 1325 as an important act of the Council regarding peace operations,Footnote 41 and it identifies women’s special safety, hygiene and medical needs and pregnant women’s nutritional requirements.Footnote 42 Otherwise, similarly to the UK Manual, it largely restates the existing IHL rules regarding women’s protections.Footnote 43
The most recent version of Australia’s IHL manual available in the public domain online appears to quote the first paragraph of AP I Article 76,Footnote 44 and that women are granted “specific protection”,Footnote 45 but it does not further develop ADF views specifically related to Article 76. It does note, however, that IHL provides “inalienable legal rights to various categories of persons” including women civilians,Footnote 46 and that among non-contraband items not subject to capture by a belligerent are “clothing, bedding, essential foodstuffs, and means of shelter for the civilian population in general, and women and children in particular”.Footnote 47 Otherwise, it largely restates the existing IHL protections for women,Footnote 48 and in terms of women’s rights under occupation, it repeats the outdated formulation of protecting women against “any attack on their honour”.Footnote 49
The Canadian manual on Law of Armed Conflict at the Operational and Tactical Levels (Canadian Manual), in a paragraph entitled “Equal Treatment without Adverse Distinction”, does note that “[s]ubject to special provisions relating to … gender, protected persons must receive equal treatment”.Footnote 50 It further notes, however, that “[w]omen must be specially protected against any attack on their honour, in particular against rape, enforced prostitution or any other form of indecent assault”, essentially reverting to language like that found in GC IV Article 27 rather than restating AP I Article 76.Footnote 51 The Canadian Manual has a section that specifically discusses civilian protections under AP I, but this section does not address AP I Article 76.Footnote 52 Other descriptions of women’s protections largely restate existing IHL without elaboration.Footnote 53
The German Law of Armed Conflict Manual’s (German Manual) treatment of AP I Article 76 and the special respect to be accorded women in armed conflict is similar to that found in the IHL manuals of Australia, Canada, France and the UK. The discussion of Article 76(1) only restates the provision’s text.Footnote 54 The protections for women prisoners of war are premised on the requirement that “[w]omen must receive special respect and protection”, but the discussion of these protections consists only of a recitation of existing IHL.Footnote 55 The treatment of the protections for women internees and detainees is similar.Footnote 56
One section of the German Manual, however, does arguably reflect Germany’s recognition that certain individual civilians may be at greater risk, because of their gender, from activities that are unlawful. In its discussion of Germany’s effort to align its criminal law with the requirements of the Rome Statue of the International Criminal Court, the Manual explains the creation of the German Code of Crimes against International Law, and how this code prohibits crimes against humanity including human trafficking, particularly of women and children.Footnote 57
This lack of elaboration on what special respect means is not because these five countries are laggards when it comes to promoting gender equality. Each has played a very active role in implementing Security Council Resolution 1325 and related resolutions with National Action PlansFootnote 58 to implement this global framework. The UK is already on its fifth National Action Plan,Footnote 59 Canada,Footnote 60 FranceFootnote 61 and GermanyFootnote 62 are each on their third, and Australia has adopted two.Footnote 63 In giving practical effect to Resolution 1325, these official acts show that these nations are likely attuned to the special needs of women in armed conflict, and are also then evidence of State practice consistent with a broader application of the special respect for women set out in AP I Article 76.
In contrast to these national IHL manuals, however, the manuals of two other countries with fairly recent combat experiences and whose societies are recognized internationally for their very progressive approaches to gender equality, Norway and Denmark,Footnote 64 show a more comprehensive approach to the special protections afforded to women. Admittedly, although both manuals cite AP I Article 76 in their discussion of the IHL protections afforded to women,Footnote 65 neither explicitly address the meaning of Article 76(1)’s special respect. Instead, these manuals provide detailed explanations and practical explanations of special protections to be accorded to women that in effect expand the application of special respect beyond the text of Article 76.
The Norwegian Manual of the Law of Armed Conflict (Norwegian Manual)Footnote 66 recognizes that women and men receive equal protection under IHL, but that “[i]n practice, however, women and children are more vulnerable to suffering in wartime. Among other things, they are more susceptible to assault and inhumane treatment.”Footnote 67 In this vein, although it restates the GC IV Article 27 formulation of protecting women against sexual assault and enforced prostitution rather than AP I Article 76(1),Footnote 68 it does rely on Article 76(1) for its description of the protection accorded to women prisoners of war from sexual assault.Footnote 69 Further, it cites AP I Article 76(2) for the protections to be accorded to women detainees and internees facing judicial or disciplinary action.Footnote 70 Otherwise, it largely relies on the Geneva Conventions for its description of protections afforded to women.Footnote 71
In this sense, the Norwegian Manual hews closely to the national manuals already reviewed. It departs from them, though, in its thorough explanation of the rationale for the special protections accorded to women. It clarifies, importantly, that while humane treatment of civilians includes making no adverse distinction based on sex, adverse distinction means “discrimination that is biased or unreasonable, such as giving better medical treatment to a specific ethnic group without it being medically justified”.Footnote 72 This is coupled with the explanation that because “various groups have been found to be particularly vulnerable to suffering during war, rules have been developed to ensure that account is taken of their particular needs”,Footnote 73 including women.Footnote 74
The Norwegian Manual also departs from the other manuals reviewed above in the way it uses practical examples to describe what the special protections for women look like in application. For example, it explains that from a practical perspective, different cultures have different views on the interaction of men and women than does Norway, and that part of affording special respect to women in these places means planning for women soldiers to conduct searches of local women.Footnote 75 It also notes that it is important to establish hospital and safety zones in order to “protect persons who are particularly vulnerable to the effects of war”, including pregnant women and mothers with young children under 7 years of age.Footnote 76
Couched in terms of assessing proportionality rather than explicit protections for women, the Norwegian Manual also contains targeting scenarios that are premised on civilian women and children being present at all times in dwellings located near a military objective.Footnote 77 This presents women and dependent childrenFootnote 78 as specific categories of civilians to be considered in targeting decisions, rather than just assessing the impacts on civilians generally. Similarly, the scope of the proportionality analysis set out in the Norwegian Manual regarding damage or destruction of civilian objects is described in a way that lends itself to the needs of individual civilians. The Manual notes that whilst inconvenience to the civilian population through a reduced selection of food is not collateral damage to be considered, “a lack of access to food in general would be damage that must be incorporated into the proportionality assessment”.Footnote 79 This arguably reflects protections for pregnant women and mothers of infants regarding their nutritional requirements which are not stated in AP I Articles 51 and 57.
Similarly, the Military Manual on International Law Relevant to Danish Armed Forces in International Operations (Danish Manual) emphasizes the protection of women through the prohibition on discrimination on the basis of sex, and the prevention of sexual violence against women, including forced pregnancy and forced sterilization.Footnote 80 The Danish Manual also notes that Denmark is party to many international human rights conventions, some of which focus on affording certain groups of people special protection, such as the UN Convention on the Elimination of All Forms of Discrimination against Women (CEDAW).Footnote 81 Importantly, though, as Tarasevich has argued for application of CEDAW Article 5(a)’s requirement for elimination of gender stereotyping in military targeting, the CEDAW also addresses equality of treatment for both men and women in this regard.Footnote 82
To complement the formal equality between men and women in IHL, the Danish Manual states that women have “special protection needs which must be respected”, and it restates AP I Article 76(1).Footnote 83 It notes that the prohibition against sexual assault is a special protection for women under Article 76(1)Footnote 84 because women “are often particularly vulnerable during armed conflict”.Footnote 85 The Manual also notes that there are specific categories of women who are provided special protection, such as pregnant women and mothers with infants, citing the subsequent paragraphs of Article 76.Footnote 86
Importantly, the Danish Manual goes further and explains that Danish forces are required to “effectively protect women against violence” by separating them from men while they are detained; that in areas where civilians are being detained, guards should be given guidelines that address “dignity and integrity issues”; and that in common areas, “surveillance systems should be installed”.Footnote 87 Finally, it notes that in occupied areas, this responsibility to protect might even “involve taking a closer look at the legislation of the territorial State to ensure that the protection of women is duly reflected in its domestic laws”.Footnote 88 This is a very practical approach to getting military personnel to think concretely about fully ensuring women’s protection in armed conflict.
Although it is not identified as relating to special respect for women explicitly, the Danish Manual provides an expansive definition of protected civilian objects from a targeting perspective. It first notes that civilian objects which are indispensable to the survival of the civilian population include “food, … drinking water installations and supplies, and irrigation works”,Footnote 89 which are already listed in AP I Article 54.Footnote 90 It further notes that depending on geographical and weather conditions, they could also include “medicine, electricity, gas, clothes” and blankets.Footnote 91 Although the Danish Manual does not explicitly link these considerations with the special needs of pregnant women and women with small children, all of these things could be particularly important for such women because of the heightened risks they face due to armed conflict.Footnote 92
Summary
Although the IHL manuals of these seven States Parties represent a small sample of State practice, they are particularly relevant. Each country ranks in the top thirty-five countries for gender equality worldwide, with Australia, Denmark, Germany, Norway and the UK being in the top fifteen,Footnote 93 so the fact that none of these manuals makes a direct link between Article 76(1)’s special respect for women and the conduct of hostilities is not due to indifference to gendered impacts upon women.
Whilst the IHL manuals of Australia, Canada, France, Germany and the UK do not go far beyond the text of AP I Article 76 in explaining or applying the special respect to be accorded to women, these manuals likewise do not show any explicit efforts to confine the interpretation of it to just the article’s specific provisions. Similarly, the Norwegian and Danish IHL manuals do not explicitly explain the meaning of this special respect in their treatments of Article 76. These manuals do, however, contain significant content that implicitly fleshes out what gender-differentiated protections for women mean, and most importantly how they could be applied. In this regard, the Danish Manual is arguably the most comprehensive of the seven manuals reviewed.
Whilst IHL manuals are significant instances of practice-oriented guides for military personnel,Footnote 94 other military guidelines are also important to consider. The doctrine that militaries develop to conduct their operations in the presence of civilians in general, civil–military cooperation and human security doctrine, are important in this regard, and there are significant instances of these guidance documents addressing the gender-differentiated impacts of military operations and armed conflict upon women.Footnote 95 There are concerns, however, about the scope of some of these documents’ coverage and their consistency with overarching multinational efforts in the area of human security.Footnote 96
The doctrinal area that poses the greatest challenge to the meaningful inclusion of gender-differentiated impacts is arguably the application of kinetic force in the presence of civilians in armed conflict. There are likely a number of reasons for this, including the lack of institutionalization of lessons learned from gender-aware military activities such as the use of femaleFootnote 97 or mixed-genderFootnote 98 engagement teams and the lack of practicable theory of gender’s operational relevance that works well across the range of different military missions.Footnote 99 Thus, it is particularly useful to review two instances of State doctrinal practice that have sought to include the gender-differentiated impacts of armed conflict in the conduct of kinetic operations – an Australian effort regarding air operations, and the current North Atlantic Treaty Organization (NATO) doctrine on targeting – even though neither explicitly addresses the special respect to be accorded to women under AP I Article 76(1).
Australian Defence Force: Pre-doctrinal work
The ADF has been a leader in factoring gender considerations into military operations.Footnote 100 Its gender advisers (GENADs) have served in conflict zones such as Afghanistan and Iraq, and the ADF has recognized their accomplishments.Footnote 101 Importantly, ADF GENADs are reported to have been creative and persistent in finding functional positions within deployed headquarters that have allowed them to meaningfully contribute to the planning of both kinetic and non-kinetic operations.Footnote 102
Consistent with the ADF’s recognition of the need to consider gender in operations, in 2018 it published an innovative pre-doctrinal document, Air Force Doctrine Note 1-18, Gender in Air Operations (AFDN 1-18).Footnote 103 It is important to note that ADF doctrinal notes are not doctrine themselves, but they are intended to address “specific doctrinal matters that need to be formally articulated between major doctrinal reviews” of higher-level capstone doctrine; they are designed “to inform and promote discussion on a specific doctrine subject and may not necessarily represent an agreed position”.Footnote 104 Doctrinal notes are temporary.Footnote 105
Of particular relevance to the topic of gender differentiation in precautions in military operations, AFDN 1-18 provided a hypothetical example of an attack on a bridge to explain the differences between immediate and subsequent effects related to gender that should be considered in targeting.Footnote 106 If, for example, an enemy force were to use the bridge to transport materiel, then destroying the bridge could provide an attacking force with a lawful military advantage by complicating the enemy’s logistical efforts. This would be an immediate or first-order effect.Footnote 107 If, however, local civilian women used the bridge to access essential supplies, the destruction of the bridge could result in gendered second-order effects, such as forcing the women to use less secure routes to gather these supplies.Footnote 108 Use of less secure routes could lead to significant third-order effects, such a foreseeable increased risk of sexual or gender-based violence against the women.Footnote 109 If such sexual or gender-based violence is foreseeable, then it would be within the harms to individual civilians covered by AP I Article 57.
AP I Article 57 requires that feasible precautions be taken to minimize incidental injuries among civilians expected to be affected by an attack, and although it did not explicitly link Article 57 to targeting, AFDN 1-18 reasoned that “[a]pplying a gender perspective in analyzing information and assessing intelligence allows greater clarity of the direct and indirect effects of targeting”.Footnote 110 Assuming the data and analysis of the use of the bridge could be developed sufficiently to establish the foreseeability of sexual violence against women, this information would then arguably be directly applicable to compliance with Article 57(1)’s requirement for militaries to exercise constant care towards civilians in operations, and Article 57(2)’s requirement to take feasible precautions.
It is unclear whether concepts from AFDN 1-18 have been meaningfully incorporated into official ADF doctrine guiding air operations. For example, the most recent ADF doctrine regarding air operations available in the public domain does not cite the doctrine note as a resource, and its treatment of gender is cursory.Footnote 111 Additionally, it is important to note a limitation in AFDN 1-18’s construction that might have lessened its impact upon kinetic air operations doctrine development. Although it identified the need to take precautions in attack, it did not go further and address whether the second- and third-order gendered harms it identified could be seen as sufficiently foreseeable such that they should be included in a proportionality analysis of an attack on the target bridge.Footnote 112
The closest AFDN 1-18 came to grappling with the role of gender-differentiated harms in this context was to note that in the context of an effects-based approach to operations, “collateral damage or unintended civilian death occurring because of [sic] Air Force employing kinetic weapons could prove to be highly counterproductive to the aims and objectives sought by the Government in the conflict”.Footnote 113 Thus, “[k]inetic & non-kinetic effects which consider gender experiences are more effective in realising strategic outcomes, as opposed to tactical wins”.Footnote 114 This conclusion appears to have been grounded in considerations of practical impacts – which might be accurate – and not explicitly linked to IHL. NATO’s joint targeting doctrine, however, does consider the full range of legal requirements regarding attacks, as well as the role that considerations of gendered harms could play in conducting attacks.
NATO doctrine
NATO published Allied Joint Publication 3.9, Allied Joint Doctrine for Joint Targeting (AJP 3.9), its updated targeting doctrine, three years after AFDN 1-18 was issued.Footnote 115 AJP 3.9 emphasizes the importance of ensuring a full legal analysis of any engagement prior to its start, addressing whether there is military necessity for the attack, ensuring that combatant or military targets are distinguished from civilians and civilian objects, taking necessary precautions to minimize harm to civilians, and assessing the proportionality of the attack.Footnote 116 Importantly, this document also includes substantive provisions intended to bring gender factors into consideration when planning and executing kinetic operations.Footnote 117
“Joint targeting” in NATO is defined as “taking actions in one or more of the operational domains, using all capabilities available, against a target, in order to create an effect in one or more of the physical, virtual, or cognitive dimensions.”Footnote 118 Land, air and cyberspace are examples of operational domains, and an effect can be kinetic, such as the destruction of an enemy weapon system, or non-kinetic, such as an information operation causing a change in attitude on the part of local civilians.Footnote 119 In the target development phase under AJP 3.9, designated GENADs would provide a gender analysis of the target,Footnote 120 and in the subsequent assessment phase, the different effects of the action on women, men, girls and boys would be evaluated.Footnote 121 AJP 3.9’s operational rationale for this work is that the “integration of a Gender perspective contributes to the orchestration of fighting power”.Footnote 122
Consistent with IHL, AJP 3.9 “recognizes that all feasible precautions must be taken to avoid, minimize and mitigate harm to civilians” and that when implementing measures to protect civilians, “NATO should consider those groups most vulnerable to violence within the local context”.Footnote 123 Consistent with NATO’s Protection of Civilians Policy,Footnote 124 this includes women.Footnote 125 This provides space within the doctrine for an application of AP I Article 76(1)’s special respect for women in taking steps to protect the civilian population and individual civilians from the dangers of military operations under AP I Article 51(1), and in implementing the duty of constant care to spare the civilian population, individual civilians and civilian objects under AP I Article 57(1).
There are, however, practical challenges in making the joint targeting process work efficiently in its consideration of gender-differentiated impacts. The first issue is the recency of the GENAD position in generalFootnote 126 and the relative lack of institutionalization of the position as compared to, for example, legal advisers (LEGADs). AP I carves out a role for LEGADs in providing IHL advice to military commanders,Footnote 127 and even nations that have not ratified AP I, such as the United States, agree that this requirement has become customary international law.Footnote 128 Contrasting the respective roles of GENADs and LEGADs, Durhin, a senior NATO operational LEGAD, points out that unlike GENADSs, LEGADs will likely be present at all relevant staff meetings in the targeting process, they will have received substantial education and training on IHL, and they will have internationally recognized standards that they can apply to issues throughout the targeting process.Footnote 129
Durhin recognizes that a GENAD can provide a valuable staff service by generating a gender analysis of an operational area and explaining how it would apply to operations,Footnote 130 but notes that it will be the LEGAD’s role to explain how IHL, including special protections for women, applies.Footnote 131 This is of course the LEGAD’s job, and it is particularly the case in providing advice to commanders on applying the principle of proportionality. However, in the areas of minimizing harm to civilians in general and taking precautions in attack, including steps that a military organization could take to regulate military activities that are not directly related to the application of kinetic force, the LEGAD’s role is not likely to be so pertinent compared to other staff officers with expertise in these matters. GENADs, members of other staff sections and other special advisers might be better suited to assist with this analysis within their own functionality in order to ensure that gender factors are considered across the command.Footnote 132
This is demonstrated well in NATO’s Protection of Civilians handbook.Footnote 133 The handbook contains extensive descriptions of steps that can be taken to better effect protection for women during armed conflict, though generally in a non-kinetic context.Footnote 134 These steps are largely within the missions of different military units with different military specialties other than LEGADs, although the handbook does note the legal background for conducting these steps.Footnote 135 However, if it is understood that AP I Article 76(1)’s special respect for women is properly applicable to the implementation of AP I Articles 51 and 57 in terms of protecting civilians, this could lead to questions of IHL application that are within a LEGAD’s competency to address.
Another point that Durhin makes about the use of gender factors in targeting is that targeteers can find it difficult to satisfactorily integrate sex- and age-disaggregated data into the joint targeting process because of “technical and operational reasons”.Footnote 136 Ensuring the actionability of gender-related data in the targeting process is a long-standing concern, and military organizations would need to make a very significant investment in collecting, analyzing and norming this data in the intelligence process to make this consistently practicable.Footnote 137 Importantly, this is not just a question of targeting. If a proper reading of AP I Article 76(1) is that the special respect to be accorded to women is applicable to the provisions of AP I Articles 51 and 57, it is difficult to conceive of a military organization that could effectively meet its legal obligations in this regard without actionable gender analysis.
Finally, Durhin notes the risk that segregating civilians in the targeting process on the basis of gender might cause military staffs to not protect all civilians equally. Planners and operators might be “misled and become tempted to assess possible collateral damage as ‘more acceptable’ if the victims are exclusively male”.Footnote 138 As Fluhr has pointed out, there is evidence that Western military organizations during the war in Afghanistan showed bias in targeting people as taking direct part in hostilities on the basis of their presumed maleness.Footnote 139 Tarasevich notes that the greater the role that qualitative and context-dependent factors play in targeting assessments, the greater is the risk that these judgements will be influenced by gender stereotypes.Footnote 140 Durhin’s point is therefore very important, because it goes right to the heart of the argument as to why the ICRC should consider incorporating gender factors in the revised Commentary to AP I Articles 51 and 57: ensuring equality of treatment of women and men.
Summary
The ADF’s AFDN 1-18 and NATO’s AJP 3.9 are two examples of doctrinal work that seeks to bridge the gap between the explicit special protections set out for women in IHL and the protections afforded the civilian population and individual civilians in AP I Articles 51 and 57 – though they might be the only two. Taken together with the approaches to the gender-differentiated impacts of armed conflict shown by the IHL manuals of Norway and Denmark in particular, they are substantive evidence of positive State practice to make women the object of special respect, even if they do not cite AP I Article 76(1) explicitly. Thus, analysis under VCLT Article 31 supports the conclusion that the special respect to be accorded to women under AP I Article 76(1) should not be read as limited to the instances of special protection within that article, and should instead be seen as applicable to the whole of AP I. To be more certain that it would be proper to read the special respect for women into the provisions of AP I Articles 51 and 57, as applicable, it is useful to briefly review the negotiation history of Article 76.
VCLT: Supplementary means of interpretation – negotiation histories
The first conferences of Red Cross and government experts
The four different sessions of the Diplomatic Conference which resulted in the final version of AP I were preceded by successive conferences of Red Cross experts and government experts. At the first conference of Red Cross experts held at the Hague in March 1971, an ICRC representative addressed the question of protection of the civilian population in armed conflict, and set out four special points for the experts’ consideration, including “the special protection of certain categories of the civilian population”.Footnote 141 The conference experts divided those who should receive special protection into two major groupings; among the first were persons considered weak and vulnerable, and among the second were those who would go to their aid, such as civil defence workers.Footnote 142 As to those who were vulnerable to the dangers of armed conflict, the experts saw these as including “[w]omen in special condition: (expectant mothers, maternity cases, mothers of infants)”.Footnote 143
Similarly, at the first conference of government experts, held in Geneva in late May and early June 1971, the ICRC asked the experts whether certain categories of persons should be given special protection in addition to the general protection they already enjoyed as part of the civilian population,Footnote 144 and whether this could be done without compromising their general protection.Footnote 145 The ICRC noted that provisions of GC IV provided enhanced protections, such as for expectant mothers and maternity cases under, for example, Articles 16, 18 and 23,Footnote 146 and that the GC IV Commentary on these articles explained that this special protection was “superposed” on the general protection accorded civilians.Footnote 147 Accordingly, the ICRC found that
[t]he existence of norms of special protection has the additional advantage of allowing certain civilians to be favoured, in an objective manner, by having recourse to criteria to which reference has already been made: state of persons (age, sex, weakened physical or mental condition, etc.) and their functions (aid and help to the aforementioned persons).Footnote 148
Although the report of this conference identifies sex as being one of these objective states, this does not mean necessarily being female per se. Instead, the special protection to be given to women in GC IV is often keyed towards their observable reproductive status, because of the increased risks from armed conflict that they face in childbearing, childbirth and care of young children. Unless these physical indicators of heightened risk are observed and tracked, it would be difficult to accord women in these circumstances special protection. This suggests that the collection of specific data would be necessary to comply with these special protections.
The second conferences of Red Cross and government experts
The second conference of Red Cross experts was held in Vienna a year later, and by this point in time, the ICRC had prepared a draft of the articles to be included in AP I for the experts to consider.Footnote 149 In his address at the opening session of the conference, the Austrian minister for foreign affairs stated that the suffering experienced in armed conflict called “not only for a more liberal interpretation, but for a broader application of the Geneva Conventions, whose sole purpose is to help people in a thoroughly impartial manner and, in particular, to protect women and children”.Footnote 150 Whilst special protection for children was both a topic of discussion amongst the experts and factored into proposals put forward by different National Red Cross and Red Crescent Societies,Footnote 151 special protection for women was not addressed. Similarly, at the second conference of government experts convened in Geneva shortly thereafter, between 3 May and 3 June 1972, there does not appear to have been discussion on any additional practical measures that parties to conflicts might undertake to implement the special protection and respect to be accorded to pregnant women and maternity cases.Footnote 152
By the time of the 22nd International Conference of the Red Cross and Red Crescent (International Conference) in Tehran in 1973, the structure of the draft articles had been rearranged slightly. At this point, pregnant women and maternity cases were defined as being amongst the “sick and wounded”,Footnote 153 and other measures regarding the special protection for women were now collected in a new draft Article 67, which stated that “[w]omen shall be the object of special respect and shall be protected, in particular against rape, enforced prostitution, and any other form of indecent assault”, and that the death penalty could not be executed on pregnant women.Footnote 154
The Diplomatic Conference
The draft version of AP I provided to the 22nd International Conference in Tehran in 1973Footnote 155 also served as the starting point for the Diplomatic Conference which began in early 1974.Footnote 156 Draft Article 46, “Protection of the Civilian Population”, continued the thread of considering both the general civilian population as a whole and individual civilians that had begun at the first conference of Red Cross experts in 1971, stating that “[t]he civilian population as such, as well as individual civilians, shall not be made the object of attack”.Footnote 157 This differentiation, which is important to the implementation of special protection for groups of people particularly at risk to the dangers of armed conflict such as women, does not appear to have been controversial among the national delegations, and none of the many amendments to this article that were offered by different nations addressed this phrasing.Footnote 158
By the end of the fourth and final session of the Diplomatic Conference in 1977, draft Article 67 had changed into what became AP I Article 76, “Protection of Women”.Footnote 159 In introducing the draft article to the national delegations at the committee considering it, the ICRC representative noted that the intent of this provision in part was to expand the protection against sexual assault provided to women in GC IV Article 27 to all women. Further, it was intended to provide special protection for pregnant women, maternity cases and those caring for young children.Footnote 160
For the most part, the amendments offered by the national delegations and the discussions of them concerned the pronouncement of death sentences against pregnant women and those with young children;Footnote 161 there was no significant disagreement on according special protection to pregnant women, maternity cases and those caring for young children. Likewise, there were only a couple of national proposals to broaden the scope of the special protection to be afforded women in armed conflict – for example, Vietnam proposed “a special safeguard for pregnant women arrested for patriotism or political non-submission”.Footnote 162 In the end, the national delegations did agree that pregnant women and mothers with dependent infants whose liberty was restricted for reasons related to an armed conflict would receive the “utmost priority” in having their cases considered.Footnote 163
Summary
A review of the negotiation history of AP I Article 76 shows that the different groups of Red Cross and government experts working on its formulation were open to finding new ways to implement its provision of special respect for women. In the end, Article 76 built on the existing protections afforded women under the Geneva Conventions in substantive ways, but it did not really explain the reach of the “special respect” for women that it established. Further, the use of the term “individual civilians” in addition to the general civilian population does not appear to have been controversial at all during the AP I negotiations,Footnote 164 suggesting that the ICRC’s rationale for its use was understood and accepted.Footnote 165 Finally, although the negotiation history does not indicate that the different groups of experts discussed applying the special respect for women in Article 76(1) to Articles 51 and 57, there is likewise no evidence that this special respect was intended to be limited to only those provisions contained in Article 76 itself.Footnote 166
Protecting civilians from armed conflict’s effects: Articles 51 and 57
AP I Articles 51 and 57 represent important instances of AP I’s combination of the “Geneva Law”, which protects victims of armed conflict who have fallen into the power of the enemy, with the “Hague Law”, which regulates the conduct of armed conflict, into one treaty.Footnote 167 The AP Commentary’s interpretations of these articles are promising sites for revisions aimed at including the gendered effects of armed conflict upon women and girls, because as Gardam and Jarvis have noted, “IHL fails to take into account that the application of the provisions of the Law of the Hague determining lawful targets in armed conflict has gendered consequences”.Footnote 168 A brief review of these provisions shows that including AP I Article 76’s special respect for women in applications of Articles 51 and 57 would be consistent with the latter articles’ plain language and related purposes.
General protections and precautions in attack
AP I Article 51 provides that the civilian population and individual civilians “shall enjoy general protection against dangers arising from military operations”Footnote 169 and sets out a non-exhaustive list of rules to this effect, including the prohibition of reprisals against them.Footnote 170 AP I Article 57 likewise focuses on the impacts of military operations, but rather than being cast in terms of a right that civilians have, it instead looks at the precautions that States Parties must undertake in conducting attacks from the perspective of a duty of constant care “to spare the civilian population, civilians and civilian objects”.Footnote 171 Importantly, the operations covered by these articles are largely the same from a practical perspective, and the articles themselves do not restrict covered military operations to just the application of kinetic force against targets.Footnote 172 Articles 51 and 57 cover a broader range of military undertakings that could cause civilian injuries or deaths, or damage to civilian objects.Footnote 173
This broader range of covered military activity, read together with the general protection afforded to the civilian population and individual civilians under Article 51 and the constant care to spare them harm under Article 57, suggests that holistic and granular data about civilians in general and different categories of individual civilians would be necessary for military organizations to know whether they were complying with these requirements.Footnote 174 The special respect to be shown to women under AP I Article 76(1) provides a legal basis for identifying the collection and analysis of gender-disaggregated data as a necessary part of these efforts.
Further, among the specific precautions in attack required of States Parties under Article 57 is the obligation to “take all feasible precautions” in the choice of means and methods of attacks “to avoid[ ], and in any event to minimiz[e]”, civilian injuries and deaths, “and damage to civilian objects”.Footnote 175 Another precautionary requirement is that those who plan or decide upon attacks first do “everything feasible” to confirm that the objectives of the attack “are neither civilians nor civilian objects and are not subject to special protection”.Footnote 176
The scope of the term “everything feasible” was discussed at length during the Diplomatic Conference, and the national delegations chose this formulation over the ICRC’s proposed language of “take all reasonable steps”. That phrasing was seen as too broad and as being beyond the senses of practicability for which the negotiators were aiming.Footnote 177 As to the level of effort needed to meet this standard, the AP Commentary notes that this could, for example, require planners to request additional reconnaissance information on potential targets, as well as “a serious check of [that information’s] accuracy”.Footnote 178 In a more recent assessment of the effort required, the committee established by the International Criminal Tribunal for the former Yugoslavia (ICTY) to review NATO’s bombing campaign in Serbia and Kosovo found that “[a] military commander must set up an effective intelligence gathering system to collect and evaluate information concerning potential targets”.Footnote 179
In describing the role of intelligence in complying with IHL, Voss, a US Air Force intelligence officer, explains that it is not enough to identify a target’s significance to enemy operations. In addition, intelligence must “provide a deep understanding of the civilian aspects of the operating environment that will be foreseeably impacted by the target’s engagement”.Footnote 180 Voss helpfully translates the requirement for accurate and effective intelligence gathering and analysis into good practices that military planners and operators can apply irrespective of the specific military organization in which they are working.Footnote 181 A number of these practices could be particularly useful in capturing the gender-differentiated harms caused by military operations, such as collecting intelligence “of a different discipline that was previously collected”,Footnote 182 feeding data on the effects of previous attacks into the analysis of potential future attacks “to ensure that compounding effects are understood”,Footnote 183 and using intelligence to “model and predict internally displaced persons and refugee flows”.Footnote 184
Thus, if Article 76(1)’s special respect for women provides a legal requirement for including gendered harms to civilians amongst the body of data properly considered in the application of Article 57 by military organizations to their operations, the notion of feasibility suggests that they must make a good-faith effort to develop that information for consideration.Footnote 185 Of particular interest here is the additional requirement to determine that the objectives of an attack “are neither civilians nor civilian objects and are not subject to special protection”.Footnote 186 Although the applicability of the term “special protection” was not discussed at the Diplomatic Conference,Footnote 187 even as early as the first conference of government experts held in 1971 there was extensive discussion of “special protection” in the context of categories of civilians who would benefit from differentiated treatment as compared to the civilian population in general.Footnote 188 In particular, even in the context of special protection for certain non-military objects, the special protection due to certain categories of civilians was noted.Footnote 189
The AP Commentary only notes that the text of Article 57 requires an assessment by those launching attacks of whether there are objects subject to special protection, and “in particular whether they are cultural objects or places of worship”,Footnote 190 but the text itself does not explicitly confine special protection to civilian objects. More recently, Corn and Culliver have seen assessment of both civilians and civilian objects as being required in terms of any special protection status review.Footnote 191 Article 76(1)’s special respect for women would therefore provide a legal basis for the consideration of gendered harms as part of this assessment as well.
Developing an effective system for the collection and analysis of gender-disaggregated data which is keyed to objectively recognizing the gendered risks that women face as they experience armed conflict poses certain challenges. As Jarvis and Gardam point out, meaningful incorporation of gender factors is not simply a matter of counting women and girls, because to talk only “in terms of sex or gender disaggregated civilian casualties, as is the usual practice, in so many cases obscures and renders invisible the particular effects of warfare on women and girls”.Footnote 192 These challenges are heightened when considering the rule of proportionality set out in Articles 51 and 57, and the foreseeability of gendered harms in the context of what are called “reverberating effects.”
Proportionality and the foreseeability of reverberating effects
Both Articles 51 and 57 of AP I state a rule of proportionality prohibiting attacks that are expected to cause incidental injuries to civilians or civilian deaths, or damage to civilian objects, which would be excessive in relation to the “concrete and direct military advantage” anticipated by the attacking State Party.Footnote 193 In determining whether the attacks they intend to launch meet this standard, commanders must take a broad approach in considering information relevant to incidental harms that are caused by the attacks. The United States has not ratified AP I, but it has long agreed that there is a customary international law rule of proportionality which applies to attacks,Footnote 194 and the US IHL manual notes that “[i]n determining whether the expected incidental harm would be excessive, the totality of circumstances must be considered. This holistic judgment should consider any relevant moral, legal, and military factors.”Footnote 195 Accordingly, “[i]ncidental harm is more likely to be regarded as excessive in attacks in which it would be feasible to take additional precautions and those precautions are not taken”.Footnote 196
“Expected” means foreseeable or likely to occur.Footnote 197 This means that even effects which are not directly and immediately caused by the attack, sometimes called reverberating effects, must be considered in the proportionality analysis if they are foreseeable results of the attack.Footnote 198 Foreseeability of gendered harms can be promoted by “studies on the effects of conflicts; better modelling of weapons’ effects; … and new technologies to better assess the condition or status of infrastructure and service delivery during the conflict”.Footnote 199 Robinson and Nohle argue that “past experiences and empirical data can contribute to making certain reverberating effects foreseeable”,Footnote 200 and that if an attack which would result in displacement of civilians could foreseeably “result in increased mortality and deteriorating health of displaced persons”, then this is the sort of harm that should be included in this analysis. In the same vein, if it is reasonably foreseeable that the incidental destruction of civilian homes will lead to large-scale civilian displacement, this should be properly considered in the proportionality analysis.Footnote 201
Displacement’s health and safety consequences could be particularly pronounced for women and girls.Footnote 202 Suppose, for example, that an attack on farm dwellings being used by adversaries resulted in the displacement of families that lived there. Suppose too that this area in which the families lived was marked by pervasive and severe gender inequality, and the loss of farm dwellings meant that mothers would lose their capability to feed and to protect themselves and their children from sexual violence, and to care for old and infirm persons for whom they are responsible given the recognized cultural norms of their society.Footnote 203 These gendered harms are the sorts of risks that the special protections for women in IHL are intended to address, and they are types of injuries that precautions in attack and proportionality are intended to cover. The question is, what would make them foreseeable during military operations?
It is not clear that existing military gender analysis tools would provide sufficiently robust information for commanders, planners and operators to take action upon.Footnote 204 For example, the Military Gender Analysis Tool produced by NATO’s NCGM is one of the most recent analysis designs for including gender considerations in military operations.Footnote 205 It is intended to be flexible and holistic, and it links conceptually with the existing PMESII (Political, Military, Economy, Social, Infrastructure, Information) military analysis framework.Footnote 206 However, whilst the Military Gender Analysis Tool would be useful as a starting point in assessing the risks posed to women civilians by military operations related to hostilities, the generality of its evaluative factorsFootnote 207 would likely not provide intelligence that military organizations could immediately use to avoid or minimize gendered harms amounting to civilian injuries or deaths, or damage to civilian property.
A promising model for the type of tools that might be needed has been developed in the context of the Political Declaration on Strengthening the Protection of Civilians from the Humanitarian Consequences Arising from the Use of Explosive Weapons in Populated Areas.Footnote 208 Work done by the UN Institute for Disarmament Research (UNIDIR) has already explored in some detail what the gendered indirect impacts of these explosive weapons could be,Footnote 209 and has produced sophisticated methodologies for assessing reverberating effects that include those related to gendered harms.Footnote 210 Although it does not appear that they currently deal with gendered harms based on what information is available in the public domain,Footnote 211 different military collateral damage estimate methodologies could feasibly be adapted to include practicable portions of UNIDIR’s work.Footnote 212
Conclusion
The ICRC has recently finished its revisions of the Commentaries on the 1949 Geneva Conventions, updating them to match developments that have occurred in international law and international relations since those treaties were negotiated.Footnote 213 This work appears to have been informedFootnote 214 by the assessment of feminist legal scholars that IHL had not really recognized that women’s experiences of armed conflict are exacerbated by underlying systemic gender inequality.Footnote 215 Accordingly, the revisions to the ICRC Commentaries discarded outdated language that premised the legal protection of women on notions of their “honour” – that is, their chastity and the status of their familial relationships.Footnote 216 Instead, women’s legal protections are now recognized as being based on the plain language of the provisions of the Geneva Conventions that establish their equality with men in armed conflict situations.Footnote 217
The ICRC’s work to revise the AP Commentary is already under way.Footnote 218 This provides the ICRC with an opportunity to address the growing recognition in the international community that armed conflict often has negative impacts on women and girls as individual civilians that are different from those experienced by men and boysFootnote 219 in its interpretations of the protections afforded to civilians by AP I Articles 51 and 57.
For militaries to be able to effectively address gender-differentiated harms in operational analyses related to activities covered by Articles 51 and 57, there would need to be a significant investment in gathering and analyzing gender-disaggregated data, integrated across the breadth of the military mission.Footnote 220 These efforts would face the reality that existing collection and analysis systems likely reflect strong male-norming biases;Footnote 221 without such an investment, however, it is not clear how actionable gender intelligence could be provided to commanders, planners and operators.Footnote 222
Over time, the development of methods and means of gender-disaggregated data collection and analysis, likely leveraging artificial intelligence technologies so that these processes are both timely and affordable (as long as any gender- and age-related biases in these programs are remedied),Footnote 223 could create a state of operational awareness in military organizations of the gendered harms that their operations could foreseeably cause. As Kamunde has noted, however, even if such biases in the technology are satisfactorily addressed, “[t]he limitations of data and its access in certain conflicts, because of the protracted nature of the armed violence, could also be a basis of inbuilt biases and prejudices when employing” it.Footnote 224 Mindful of these practical risks, the ICRC’s upcoming revisions to the AP Commentary present a unique opportunity to link the development of such an operational picture with a better understanding of the legal bases for including a gender-differentiated approach into an area of armed conflict from which such approaches have largely been absent. This linkage would help us better realize the special protections for women and girls under IHL during armed conflict.
Open access
