Typologies of Violence in the Context of Disasters and Climate Change

Violence occurs in all settings—that is, in the workplace, on public transports, in the streets, in internally displaced camps, online, and within the home—but VAWG disproportionately occurs within the family or domestic unit and between former or current spouses or partners (García-Moreno et al., 2005). Also referred to as intimate partner violence (IPV), this violence is overwhelmingly borne by women in all regions of the world [United Nations Office on Drugs and Crime (UNODC), 2019; World Health Organization (WHO), 2013] and can take different forms. One typology from WHO (2005, p. 13) is built according to those committing violent acts, distinguishing self-inflicted violence from that perpetrated by family or community members (both acquaintances and strangers) and collective violence at social, economic, and political levels. Other typologies focus on the nature of violence being inflicted, and in general, four broad categories tend to be used by the majority of development and humanitarian agencies (Table 1).

More academic research seems available on the impacts of disasters on reproductive health than there is for disasters and GBV, with studies conducted after the 2008 Wenchuan earthquake in China (Liu et al., 2010), the 2004 South Asian tsunami (Carballo et al., 2005), and following Hurricanes Irma and Maria in Puerto Rico (Welton et al., 2019). A survey conducted in the Texas Gulf Coast region following Hurricane Ike in 2008 also showed that Black women had more difficulty than their White and Hispanic counterparts accessing contraception despite the presence of reproductive health clinics (Leyser-Whalen et al., 2011). However, a significantly overlooked issue is obstetric violence, which affects adolescents and women during their gynecologist monitoring, pregnancy, childbirth, and the postpartum period. In emergency settings, women’s reproductive health needs are even less likely to be adequately met (Richter & Flowers, 2008, as cited in Leyser-Whalen et al., 2011), and the violence they might continue to suffer is not documented.

The four categories of violence are not mutually exclusive, and they can be labeled differently, for example, under the umbrella terms “domestic violence” or IPV (Figure 1). One of the reasons for distinguishing different forms of violence is to support the monitoring and comparability of data [European Institute for Gender Equality (EIGE), 2017] while stressing that IPV is characterized more by a continuum of violent acts that exacerbate one another rather than a single offense.

Open in new tab

Figure 1. Typologies of gender-based violence to consider before, during, and after a crisis.

The Continuum of Violence: From Everyday Life to Crises

Studies on gender and disaster converge on the point that VAWG is a daily occurrence prior to a disaster and cases of violence post-disaster are not necessarily extraordinary events but, rather, more the continuum of a pervasive manifestation of inequality and discrimination. Discriminatory attitudes and social norms (i.e., the informal and formal laws, beliefs, and practices that help determine collective understanding of what are acceptable attitudes and behaviors; Harper et al., 2014) are major drivers of GBV (Heise & Kotsadam, 2015; Oswald-Spring, 2008). This is also one of the reasons why some forms of violence might not be perceived as such by both victims and perpetrators in the context in which they live (Rezwana & Pain, 2021). In all regions of the world, however, women and girls bear by far the greatest burden of intimate partner/family-related homicide, or rather femicide, with more than half (58%) of women intentionally killed, murdered by their partners or other family members (UNODC, 2019). Men are also more likely to be the perpetrators of homicide—out of all persons brought into formal contact with the police for intentional homicide worldwide in 2016, more than 90% were men (UNODC, 2019). Although the large majority of the literature on GBV in disasters documents VAWG, a small but increasing body of research is also examining the experiences of gender and sexual minorities prior, during, and after disasters (e.g., Dwyer & Woolf, 2018; Gaillard et al., 2017; Roeder, 2014; Roure, 2020).

Only a few studies have delved into men’s and boys’ experiences of perpetrating and suffering from GBV post-disasters (Austin, 2016; de Alwis, 2016; Molin, 2018; Nguyen & Rydstrom, 2018; Temple et al., 2011). Factors associated with men’s use of violence at the individual level include having experienced violence in childhood, work-related stress, expressing rigid gender attitudes (Barker et al., 2011), or consuming alcohol (WHO, 2013). Moreover, men who face unemployment, who own a gun (especially in the United States, South Africa, and in conflict and post-conflict settings), or who suffer mental health problems seem to be more likely to commit femicide (WHO, 2013). At the societal level, gender inequalities, including low number of women in elected government and reductions in government social spending on key sectors such as health and education, are also driving risks of femicides (WHO, 2013). All these risk factors have the potential to worsen in disaster situations.

In contrast, and although post-disaster VAWG might be the continuation of everyday domestic abuse, a few studies have shown that victims might also be women who are members of relief programs. In Japan, sexual and physical violence against women was reported in the aftermath of earthquakes in 1997 and 2011, although social pressure impeded victims’ capacities to report abuse and for any information to be adequately collected (Saito, 2012). In the case of the 2011 east Japan disaster, not only was domestic violence more frequently reported to consultation services but also young women, both evacuees and volunteers working in the evacuation centers, reported experiencing sexual harassment as well (Saito, 2012). Reports of violence occurring in emergency shelters, such as in Japan, suggest two points. First, the displacement and gathering of disaster survivors in close proximity, even temporarily, might make everyday abuse against women more known to others. Second, reports of violence inflicted upon volunteers involved in relief efforts suggest that forms of violence other than domestic abuse also take place and target women volunteers. The latter point is documented in Rachel Luft’s (2008) examination of race and gender intersectionality in the recovery phase of Hurricane Katrina in 2005. Despite the challenge of collecting information about cases of violence due to fears of reprisal, shame or stigma on behalf of survivors, as well as self-blame or confusion regarding what constitutes assault, Luft relied on participant observation during 5-month post-disaster and follow-up interviews to gather information. Her findings uncovered several dozen cases of sexual assault against volunteers of a grassroots relief organization, all but one of whom were women, plus one transgender volunteer, and White, except for one survivor. Critically, participant discourse criminalized the surrounding Black community, although almost every accused perpetrator was a nonlocal White man.

Gaps in Methods and Data

Studies on disasters, climate change, and VAWG seek to document cases of abuse that occur in crisis situations, whether existing violence is aggravated or new forms emerge. They also explore why violence takes place, the risk factors, and related changes in social structures. Quantitative analyses remain scarce due to the lack of systematic collection by governments and relief agencies of pre- and post-disaster gender-disaggregated data on GBV or simply because most incidents are unlikely to be reported to the authorities (EIGE, 2017; Rezwana & Pain, 2021). Rosborough and colleagues (2009) further stress the methodological challenges, including difficulties in accurately capturing GBV survivors’ experiences and inconsistent definitions of GBV—what constitutes abuse and according to whose definition and the variations of interpretations across settings. Cerna-Turrof et al. (2019) conducted a systematic review of the influence of disaster exposure on violence against children and reported that the heterogeneity and limitations in existing quantitative assessments prevented them from drawing firm conclusions on a positive association.

Hence, the majority of research in this domain is qualitative and takes place at the household and community level through in-depth case studies. This localized level of inquiry also responds to the objective of gender and disaster scholars to “contextualize gender”—that is, in historically and culturally specific contexts that shape social interactions which themselves reproduce and institutionalize discrimination (Enarson & Phillips, 2008). This requires methods that are sensitive to the conditions of disaster survivors, particularly when exploring issues of IPV. Often, due to ethical concerns, researchers choose not to conduct surveys directly with potential GBV survivors but, rather, collect data from grassroots organizations, service providers, and volunteers in relief programs (e.g., Fisher, 2010). Studies rely significantly on field and participant observation, in-depth interviews or life stories, focus group discussions, ethnographic research, and photovoice projects (e.g., MANTRA Project, 2017). Some studies have reflected on the processes of researching issues of GBV in post-disaster contexts (Sloand et al., 2015) or on the reliability of certain methodologies such as the “neighborhood method” to measure GBV in disaster settings based on secondary reporting (Stark et al., 2020).

Overall, studies that provide empirical evidence of long-lasting changes in power dynamics are scarce. These kinds of changes are often only uncovered through in-depth, qualitative, and longitudinal or immersive research. The scarcity of information on positive changes following a disaster might also be due to the fact that organizations that provide humanitarian assistance collect data solely through vulnerability assessments, which often do not integrate gender and power analyses. The priority of emergency programs consists of reporting what has been destroyed and lost in order to provide quickly for the needs of those affected (Fordham et al., 2006). When agencies report an improvement in the living conditions of, and relations between, community members, this is often done to note how their programs have supported improved conditions rather than to provide an analysis of changed power dynamics. The latter could also take years to unfold—far beyond the usual time frame of development projects. The more endogenous adaptive strategies developed by community members without external assistance receive far less attention (Gaillard et al., 2019). Despite the critical insights they provide, gender-sensitive assessments and the collection of sex, age, education, income, and disability-disaggregated data are not systematically conducted before or in the aftermath of a disaster [International Federation of Red Cross and Red Crescent Societies (IFRC), 2016; Morchain et al., 2015]. Yet, more gender-responsive needs assessments are critical to uncover potential coping mechanisms that trigger GBV and that could therefore undermine relief efforts. To address this gap, an increasing number of methodological resources have been developed by practitioners and researchers alike to support the assessment of GBV in humanitarian and development interventions [ActionAid International, 2014; Haiti Gender Shadow Report, 2010; IFRC, 2018b; Inter-Agency Standing Committee (IASC), 2015; Morchain & Kelsey, 2016; Sterrett, 2016].

Elusive Changes in Gender Roles and Power

Through disrupting the “normal” functioning of communities affected, disaster situations bring changes at multiple levels—that is, socioeconomic, political, and institutional. These changes may result from disaster impacts but also from the presence of relief programs. In the recovery process, the influx of humanitarian assistance means that funding becomes available to help reconstruction, from providing cash transfers to implementing protection programs. This is especially the case if disasters have destroyed much of the infrastructure and services that were initially improperly designed, creating an opportunity for “building back better” (Christoplos, 2006). Disaster recovery may open up new networks of support or pathways to previously unattainable opportunities (Bradshaw & Fordham, 2013). Because they disturb the “everyday normal,” disastrous events have been framed as open windows of opportunity to change dominant ways of thinking and acting (Birkmann et al., 2010). They can destabilize processes of reproduction of gender roles and routines, whereby affected people, particularly women, can seize opportunities to transgress gender norms—for example, by developing new skills through reconstruction work or by involvement in community-based relief projects (Enarson, 1998).

However, the nature and scale of disasters may influence the potential for such opportunities, and more research is needed to understand if sudden-onset events, for instance, are more likely to “impose” changes given the abruptness of impacts they can generate. In the case of slow-onset disasters, such as droughts in Kenya, changes in the roles of men and women in agriculture have been observed in CARE programs, with men increasingly taking on agricultural roles traditionally perceived to be women’s responsibilities, such as weeding, harvesting, and processing agricultural products (Otzelberger, 2016) or collecting water and fuel despite this being originally perceived as in the domain of women and girls (Webb, 2016). However, to what extent changes in gender roles tackle discriminatory gendered norms, reduce women’s unpaid household work, or do not lead to backlash is underdocumented. One study in Tanzania revealed that if men increasingly take on the role of fetching water, particularly during drought periods, they link this task to the cash economy by selling water to businesses and households (Van Aelst, 2016). Therefore, the perceived changes in division of labor and “productive” use of water do not structurally transform gendered norms and the way women’s traditional tasks are (not) valued. Gender analyses typically ask whether men taking on women’s traditional roles benefits women’s conditions and positions—whether it reduces women’s workload and improves their control over resources and their power in decision-making processes, which is why they are so important for relief and recovery processes (Morchain et al., 2015).

Such analyses, which are increasing in DRR programming and disaster or climate-related research, allow for a better understanding of inequalities in gender roles that remain or even worsen after a disaster (Fordham et al., 2006). For instance, following the Great East Japan Earthquake of 2011, women in the evacuation centers were required to prepare meals for which they received no compensation. Male evacuees, on the other hand, were not expected to contribute to this task and had the option of collecting and removing rubbish, for which they received compensation (Saito, 2012). The reliance on familial support after a crisis may also reinforce established power relations (Hoffman, 1998, as cited in Bradshaw & Fordham, 2013). Gender analyses also discuss differences of perceptions of what changed. For example, in Nicaragua, many women believed their status had changed after Hurricane Mitch in 1998, but their male counterparts disagreed; Bradshaw and Linneker (2009) examined what this indicated about the women’s position in their community.

Overall, women, girls, and sexual and gender minorities experience a re-emphasis of their often lower household status and community position in times of emergencies and also a worsening of their exposure to GBV. In Haiti, women living in camps after the earthquake in 2010 had to cope with male-dominated committees controlling aid distribution, which often forced women to negotiate through the use of sexual favors in order to meet basic needs and obtain access to supplies (Horton, 2012). In Fiji, Dwyer and Woolf (2018) documented experiences of sexual and gender minorities following tropical cyclone Winston in 2016. Not only did sexual and gender minorities face violence and harassment on a daily basis but also they were subject to blame by other community members for bringing Winston to Fiji as God’s punishment for their sexual orientation, gender identity, or gender expression. Moreover, CARE’s programming in Ethiopia reported the influence of gender norms on food distribution within households, where men and children younger than age 3 years were supposed to eat first, followed by boys and girls, and finally women (CARE Ethiopia, 2016, p. 3).

Stressors and Mechanisms That Reinforce Gender-Based Violence

Increased Displacement, Migration, and Violence

Migration, both seasonal and permanent, is a coping strategy for many people affected by disasters and environmental changes. It can be a life-saving mechanism to flee danger or violence, and it can provide new sources of livelihoods, particularly for men, who tend to rely on this strategy the most because their mobility is not restricted by gender norms. However, patterns of displacement depend on the nature of hazards, the location and socioeconomic characteristics of affected communities, and the gender divide (Millock, 2015). Issues of GBV in relation to migration can occur while people are on the move but also in displaced settings and camps or in the place of origin where migrants have left family members behind.

Experiences of internally displaced peoples (IDPs) and refugees and the protection issues that they face in temporary housing or shelters tend to be covered by a large body of knowledge because displacement also occurs as a result of armed conflicts and poverty, not just natural-hazards related disasters. Situations of high social and economic stress found in temporary housing and shelters compound pressures on displaced people who have often witnessed a disaster, survived attacks, and experienced the trauma of displacement. The combination of these risk factors underpins increased levels of GBV in displaced settings. In addition, cramped living circumstances, poor lighting and lack of security at water and sanitation facilities (Felten-Biermann, 2006; Rashid & Michaud, 2000), and the dearth of channels in which to report violence, sexual exploitation, and cases of early marriage leave women and girls at higher risk of abuse (Asgary et al., 2013; CARE, 2014; IFRC, 2012; Jayasinghe, 2015; Standing et al., 2016). In Haiti in 2010, at least 242 cases of rape against women were recorded in relief camps during the first 150 days following the earthquake (Amnesty International, 2011). In Afghanistan, Majidi and Hennion (2014) found that 64% of surveyed women reported domestic violence during displacement, often as a result of increased stress experienced by husbands. Displaced women and girls faced a higher risk of forced prostitution and forced marriage; 27% of girls were forced to marry against their will, and widows were frequently forced to marry a relative of their late husband. Daughters were also targeted for low-cost marriage by outsiders taking advantage of IDPs. IDPs are therefore entrapped in situations of acute psychological or physical violence with very little information on ways to escape (Majidi & Hennion, 2014). Fisher (2010) also highlighted the issue of long-term abuse: Although opportunistic perpetrators of sexual violence took advantage of the initial chaos following the 2004 tsunami in Sri Lanka, domestic violence continued during rehabilitation and reconstruction for several years. In contrast, earthquake-related internal displacement in Haiti did not appear as a driver of long-term violence against children (Cerna-Turoff et al., 2020).

The scarcity of natural resources, particularly in displaced settings characterized by high levels of poverty, complicates the tasks of girls and women, often deemed responsible for fetching water or collecting fuel. The increased demand for water in arid and semi-arid areas from displaced people in communities that rely on wells, for instance, augments queues, time spent collecting water, and risks of disputes or harassment. This is documented in studies in Mali and Burkina Faso (Le Masson et al., 2020) and in Ethiopia, where some girls’ strategies to get water at night conflict with the risk of being exposed to sexual assault (CARE Ethiopia, 2016). In a study reporting the experiences of refugees from Eritrea, Somalia, and Eastern Sudan living in camps in Ethiopia and in settlements in Uganda, Afifi and colleagues (2012) stressed the significance of other forms of violence: According to respondents, protracted conflict exacerbates the effects of climate variability and people’s vulnerability to other more acute perceived risks, including looting of crops by government troops, taking of children for military service, or violence toward family members and communities.

Finally, the post-disaster migration of men documented across many contexts (Enarson, 2000) means that other household members who stay behind can experience an increase in their workload and are forced to look for new opportunities to generate an income even though they are less mobile and less able to migrate outside the impacted area (CARE Ethiopia, 2016). In the Borana pastoral region of Ethiopia, some women reported benefitting from greater mobility and greater control over income due to new non-livestock livelihood activities, but their workloads also increased and they reported observing less solidarity between community members (Ridgewell et al., 2007). Furthermore, research in Chad showed that economic violence against women prevails even when men migrate: Gender norms and controlling behaviors dictate that women should not engage in income-generating opportunities, even when men remarry elsewhere or do not return (Le Masson et al., 2019).

The Failure of Protective Systems

Deaths caused by disasters and displacement increase the number of widows and orphans, result in the loss of support networks, and often shatter both formal and informal protection mechanisms (Amnesty International, 2011). While VAWG increases during emergencies, means of protection against violence diminish (FAO, 2010; IASC, 2015; IFRC, 2011, 2015). Traditional, community, or state-based support services may be disrupted, less available, and underresourced, increasing difficulty in dealing with violence (IFRC, 2016). This is particularly the case for GBV prevention and response systems, which tend not to be considered important by the public or by governments in general, let alone in the context of disasters and environmental changes. In contrast, where support for GBV survivors already exists, it is likely to resume, perhaps in different ways, after a disaster, as in the case of local shelters in New Orleans that supported both previous and additional GBV survivors (Jenkins & Phillips, 2008).

The lack of protection systems, whether disrupted by disasters or nonexistent in the first place, contributes to increase enabling environments for VAWG to occur, which is the second pathway stressed by Thurston et al. (2021). For example, the destruction of houses and resulting shortage of shelters can restrict women’s ability to leave violent relationships (Enarson, 1999). Following Hurricane Katrina in 2015, some women who were victims of domestic violence prior to the disaster had no other choice but to return to their abusive spouse due to the loss of housing, disrupted familial and social networks, and the need to access relief aid (Jenkins & Phillips, 2008). Moreover, the low social status and/or loss of income and social support of many women following a disaster make them a target of violence because perpetrators know such women are less likely to report abuse or seek assistance, especially if the legal and judicial system is not functioning (Amnesty International, 2009; CARE, 2014). In Australia, Debra Parkinson (2015)’s thesis examined the increase in domestic violence following the Black Saturday bushfires in 2009 and found that inadequate response characterized by the lack of statistics on domestic violence in the aftermath of the disaster, the neglect of this issue in recovery and reconstruction operations, and the responses to women’s reports of violence against them by legal, community and health professionals all contributed to silence violence survivors.

Discriminatory gender norms can also be replicated by relief programs (Felten-Biermann, 2006). Following Cyclone Haiyan in the Philippines, Sherwood and colleagues (2015) reported that despite many women becoming the main breadwinners for their families after losing their husbands, some organizations that conducted surveys to inform the provision of housing assistance only asked men in their communities about their needs. Even worse are cases of GBV in which sex trafficking and sexual exploitation of children have been perpetrated by humanitarian staff, security forces, or local leaders after disasters in exchange for delivering aid (Castañeda-Camey et al., 2020). Hence, although the literature shows the harmful impacts of disasters on social relations, the worsening of gender inequalities such as VAWG may increase as a negative consequence of inadequate humanitarian response rather than as a direct consequence of the disaster. In Samoa, after the tsunami of 2009 and Cyclone Evan in 2012, a study conducted by the IFRC (2016) revealed that the unequal distribution of relief supplies created disillusionment and community tensions, indirectly increasing the risk of physical violence among intimate partners. Risk of domestic violence was predominant in urban areas, where affected people had to cohabit for extended periods with host communities in crowded spaces with inadequate lighting and sanitation facilities. Young girls and adolescents were particularly vulnerable to GBV (perpetrated by male adolescents and adults) due to reduced parental supervision during the day, when parents typically left to clean up and rebuild their damaged houses (IFRC, 2016).

Therefore, whether or not the occurrence of violence increases or remains at the same level, the fact is that GBV survivors are unlikely to seek help (Gender and Disaster Pod, 2016; IFRC, 2016), protection support might be unavailable or inadequate, and the discrepancy between the prevalence of violence and the lack of protection is exacerbated in the aftermath of a disaster. Only a limited number of studies have focused on local responses to GBV—for example, the grassroots mobilization to organize self-defense or safe places against VAWG in Nepal (Standing et al., 2016) and the role of women’s refuges and social workers in attending to the needs of GBV survivors following the 2010 and 2011 Christchurch earthquakes in New Zealand (True, 2013).