Abstract
Two million people were affected in the floodplains and low-lying areas in Sirajgang in 2012. Seven hundred and fifty families were made homeless and forced to live in small temporary huts on the river protection embankments. Unemployment rose alarmingly and the jobless left their villages to find work in larger cities, leaving behind their vulnerable and insecure families. Consequently, women were increasingly required to take on totally unfamiliar roles. Our research utilised in-depth interviews with women managing without the support of their husbands. Key findings highlighted that community resilience would improve if these women were engaged at the local operational level of disaster management.
Introduction
The building of community resilience is increasingly being seen as the key to responding to the potential for disaster events in threatened communities. A community’s level of resilience is determined by the amount of both institutional and social capital on which the community’s disaster management strategies are built. Consequently, enhancing community resilience requires a building up of both. A myriad of factors determine the relative importance of each in the disaster management response. Poorer countries will tend to rely more on social capital because of the funding constraints with which their governments are confronted. However, perversely, it is generally the building up of institutional disaster response capital that is likely to have the greater impact in these countries as, in the absence of additional institutional capital, the opportunities for expanding social capital have been largely exhausted.
Exacerbating the problem of disaster in poorer countries is the socially determined role of women which, it is argued, places them at greater risk of, and more vulnerable to, disaster events. Both impacts lower the resilience of the community and call for renewed effort in building up both social and institutional capital which particularly responds to the needs of the community’s women.
Recognising this, the literature has begun to investigate the role of women in disasters – particularly in disaster prone countries such as Bangladesh, and the actions that are needed to assist them. In the next section we survey the literature relevant to the role of women in disaster where we note that an important gap is an understanding of the additional stresses placed on women whose husbands have to leave home (usually to take up work elsewhere) for extended periods (usually a minimum of six months). In Section 3 we describe our case study research method, presenting our findings in Section 4. Section 5 concludes by drawing together our findings and making recommendations.
Literature review
Flooding and other disasters in Bangladesh are increasingly being recognised as inherently gendered terrain. Bangladesh has particular challenges. It is both one of the poorest and most flood-prone areas in the world making the building of resilience, and in the context of this paper, the role of women in building that resilience, a particular concern. First, however, we need to have a definition of disaster.
Defining disaster
Disasters are defined similarly across a range of literature. Natural disasters are the main focus and United Nations (UN) and Non-Governmental Organisation (NGO) research tends to use the World Health Organisation’s definition of disaster as a ‘situation or event which overwhelms local capacity, necessitating a request to national or international level for external assistance; an unforeseen sudden event that causes great damage, destruction and human suffering’ Center for Research on the Epidemiology of Disasters (CRED), 2009, as cited in United Nations Social Development Division (SDD, 2010). This definition may be problematic for Bangladesh where seasonal flooding and its management are inextricably linked to local disaster response. While communities in Bangladesh have adapted farming practices and living spaces to cope with seasonal flooding, and even rely on it (Hanchett et al., 1998), seasonal flooding still presents serious problems. In a broad sense, seasonal flooding reduces work availability and food stores, perpetuating poverty and limiting locals’ assets and capacity to respond to larger disasters. During the rainy season, 80 percent of the labour force becomes unemployed (Brouwer et al., 2007). It also shapes farming practices in a way that further reduces the capacity to respond to disasters. Seasonal flooding, for example, has reduced the viability of certain crops, leading locals to diversify their farming methods to include shrimp farming, a practice which requires holes to be made in flood levies, increasing the likelihood of collapse in flood. Thus, coping strategies for seasonal flooding can be considered as part of the elements that must be addressed to cope with the more irregular flooding disasters. In both cases, women are at the forefront of response in Bangladesh.
Gender as an emerging focus in disaster literature
Gender is slowly gaining recognition as a research focus in the study of disaster. It is clear, particularly in Bangladesh, that the female experience of disaster is different and generally worse than that of their male counterparts and that this difference is located in the social construction of gender. A call for gender to be properly addressed in disaster literature started in the mid 1990s, with theorists noting that women were only recognised alongside other vulnerable groups such as the elderly, rather than studied in their own right (Wiest et al., 1994). A gendered disaster research agenda was articulated by Elaine Enarson in 1998 who framed it via the following key questions: 1. How is gendered vulnerability to disaster constructed? 2. How do gender relations shape the practice of disaster planning and response in household and organisations? 3. How are gender relations affected over time by the social experience of disaster? (Enarson, 1998). Since then, however, particularly in the developed world, attempts to address these research questions have been limited. In Australia, for example, no gendered disaster literature currently exists (Parkinson, 2011).
Perhaps the biggest study of Bangladesh flooding was the Flood Action Plan (FAP), which consisted of 30 studies from 1989–1995, designed by the Bangladesh government and funded globally by various governments and others. While the studies themselves are today out-dated, the politics and controversy surrounding the study represent the first major push for gendered disaster research in Bangladesh. Originally void of any gender focus, many stakeholders and NGOs claimed that the traditional methods of study did not truly capture the impact of flooding on women as distinct from their families or spouses (Hanchett et al., 1998). It was only after considerable international pressure that some gender components were added. The controversy surrounding the FAP demonstrates the way in which relief and development research programs are often intensely political and contested, shaping research in a particular way. In particular, the FAP illustrated that gender in research is a continually contested and hard-won domain. Even recently, many projects still approach gender as an ‘add on’ component, rather than an integral part of the way disasters are experienced (UN, 2009).
Women and risk
One of the key themes of disaster research is risk. Risk is becoming increasingly gender-focused. In 2009, for example, the UN called for gender-sensitive risk assessment and early warning systems, as well as gendered inter-governmental disaster risk processes (UN, 2009). If risk can be quantified or addressed it can, theoretically, be managed, or at the very least prepared for. Risk is often split into descriptions of how people protect themselves prior to disasters (preventative measures) and afterwards (coping mechanisms) (Brouwer et al., 2007). Risk is gendered not only because gender is a determining factor in risk to disaster impacts, but also because women are often in charge of assessing and reacting to risk as women’s traditional role as caregivers extends to risk management (Ariyabandu, 2003). Exacerbating the problems associated with women’s exposure to risk is the fact that, for a number of reasons, women are regarded as being more vulnerable to that risk than men.
Women and vulnerability
Women and girls are described as more vulnerable than men and more or less vulnerable than each other, depending on multiple factors. Though a central theme of the literature, there is some controversy over what vulnerability actually means, and it is often used so loosely that some have argued it is almost a meaningless word (Cannon, 2000). Some have described it as a continuum from resilience to susceptibility (Granger, 1997), but more recent literature tends to describe it as a matrix of multiple factors. According to Cannon (2000), vulnerability can be divided into five components: initial well-being, self-protection, social protection, livelihood resilience and social capital. These markers of vulnerability need to be cross-related to class, gender, ethnicity, age and political characteristics of society, so that vulnerability is made up of multiple cross-related factors. Gender has emerged as a key factor in this vulnerability matrix, as well as a key component that mediates between factors. Socioeconomic status, for example, is one of the main disaster vulnerability factors, but it is through gender-related social roles that this socioeconomic status is lived out.
Socio-economic status of women
The particular vulnerability of women to disasters and flooding is closely connected to socioeconomic status. In Bangladesh, one of the poorest countries in the world, poverty is a major factor which, added to gender, increases vulnerability. Slums tend to be in flood-prone areas and fear of losing possessions means that many people stay in these areas (Rashid, 2002). Elaine Enarson (2002) has analysed the current gendered focus of disaster literature and emphasises how social structures place some people at more risk of disaster impacts. According to Enarson, poverty places people in physical proximity to disasters and so it is the socioeconomic status of women that has the greatest effect on their ability to survive and thrive in flooding. In household interviews with 700 floodplain residents on the River Meghna in Bangladesh, it was found that the poorest people lived near the river and that those with only one income stream were the most adversely affected by flooding (Brouwer et al., 2007).
Women in Bangladesh are particularly disadvantaged as they are restricted in their work. Women earn much less than men and can only earn income in restricted ways. While the status of employed women has been improving since the 1980s, particularly with the influence of the Bangladesh Rural Advancement Committee and the Association for Social Advancement, women are still caught between two competing forces: the force of need, which expands the type of work they do, and the force of culture, which restricts it (Sultana et al., 2002). Most women are involved in fishing, but do so part-time and only to feed their family. Of the money they make from these efforts, 84 percent of women’s fishing earnings go to family expenses such as daily needs, children’s education and family health care (Sultana et al., 2002). Women still cannot undertake permanent industrial jobs and are still restricted by dowries, among other things (Ali and Niehof, 2007). Despite the increasing empowerment of women, particularly by NGO initiatives aimed at women, culture and gender roles are prevalent.
Female health
Aside from socioeconomic status, health and mortality rates are probably the biggest indicator of female vulnerability. The health of women and girls suffers more in disasters than the health of males. At the most basic level, it is widely agreed that there is a higher mortality rate for females, both during and after disaster (Parkinson, 2011). One of the problems is that Bangladesh’s health system, even in non-disaster periods, is limited (Mozumder, 2008), but flooding presents a particular problem for women, as reproductive health in particular suffers without fresh water (Ahmed et al., 2007). Health statistics are strong proof of female disaster vulnerability but they provide only a small picture of this vulnerability. For one thing, health statistics tend to be very narrowly focused on mortality and sometimes reproductive health, with very little research into other health indicators. In particular, there is almost no research into mental health, even though the psychological burden of disasters is often mentioned. More importantly, however, health statistics only provide information about the result of female vulnerability to disasters and do not explore the social and cultural factors that give rise to it. Basu and Basu (Basu and Basu, 1991), for example, found that child mortality rates are greater for working women but getting women into employment is one of the key ways of reducing their vulnerability and those of their children. Health statistics alone do not provide a proper picture of vulnerability or how the various vulnerability factors are connected. One story often mentioned in discussions of female mortality and health in flooding is of a father who, unable to keep holding on to his children in heavy floodwaters, let his daughter go in order to keep his son (Rashid, 2002). This story indicates that female mortality, and even health statistics, are merely indicative of a wider, socially-constructed, background of female vulnerability.
Violence
Another key indicator of female vulnerability to disasters is rates of violence against women, which are generally higher during and after disasters. It is generally agreed that women suffer more violence and psychological impact during disasters (see, for example, Baten and Khan, 2010; WHO, 2002). However there is very little research into precisely how or why this occurs. Some have suggested that disasters change family structures (Enarson, 2002) and heighten domestic tensions (Rashid, 2002), though it is important to note that this only explains domestic violence and not other forms of violence. Furthermore, family life is very private and difficult to collect data on (Khan, 2004).While there is some US research that indicates that domestic and sexual violence against women increases following disasters, disaster research on a global level typically does not focus on violence against women (Parkinson, 2011). According to Enarson and Morrow (Enarson and Morrow, 1998: 2), ‘disasters reveal community, regional, and global power structures, as well as power relations within intimate relationships’. One of the things we do know about women’s experience of violence is that the way societies respond to post-disaster violence tends to be based on how they perceived violence against women prior to the disaster (Parkinson, 2011) and women tend to be better off if they start jobs before marriage (Khan, 2004).
Women and existing social structures
It is often stated that disasters exacerbate existing social structures. Prior discrimination and disadvantage become magnified and result in increased hardship for certain groups during disasters relative to others. While disasters affect everyone, women confront human rights gaps during normal times and these inequalities become reified and magnified in times of disaster (Juran, 2012). Women’s vulnerability in Bangladesh, it has been argued, is due to their subordinate position in the family and embedded traditional cultural values (Rashid, 2002), but mostly the existing social structure and women’s position within this is simply assumed, rather than explored.
Vulnerability as shared but not the same
Disaster literature, and research on vulnerability in general, has begun to describe vulnerability and the vulnerability of women, as shared but not the same. Though they share important vulnerability characteristics, there is increasing recognition throughout the literature that women are not a homogenous group and that women experience disasters in different ways. In particular, certain groups of women are considered more vulnerable than others. This is an important step to examining the pre-disaster status of women as it recognises that this status is directly connected to how women experience disasters. This idea is consistent with a conception of vulnerability as a matrix of interconnected factors. It is also consistent with theoretical trends on difference generally (and gender theory specifically), which tend to evolve from a clear us-and-them dichotomy to a more nuanced recognition of internal heterogeneity and complexity. Gendered disaster theory, particularly that focused on Bangladesh and similar areas, is following this trajectory by picking up on particularly vulnerable groups within the already-vulnerable larger label of ‘female’. The vulnerable groups identified include single, young, elderly, disabled, pregnant and other such typically vulnerable groups (Wiest et al., 1994), as well as female-headed households (UN, 2009).
Women and the phases of disaster
Building resilience to disaster requires actions to be taken at all stages of the disaster cycle. While there are a number of different structures used to describe the stages of the disaster cycle, a simple one is that which splits disasters into three phases: pre-crisis, crisis and post-crisis, with the latter being divided into a recovery phase which is focused on immediate disaster recovery and the longer term adaptation which inevitably becomes preparation for the next disaster and therefore provides closure to the cycle of disaster activity (Ingham et al., 2011).
Pre-crisis
When pre-existing vulnerability is discussed, it is usually in terms of the ‘pre-crisis’ mode of disaster. Disaster research with a pre-crisis focus can be split broadly into short-term and long-term research. Short-term pre-crisis research focuses on the phase immediately preceding a disaster, where disaster is imminent, warnings are released and individuals, families and communities make choices about evacuation, food and safety. Those who face the highest risk of flooding tend to be the least prepared (Brouwer et al., 2007) and so preparation is intimately linked to socioeconomic and social status. Slums tend to be in flood-prone areas, making the already vulnerable particularly at risk (Rashid, 2002) and most people do not take preventive measures because they cannot afford to or do not know how to (Brouwer et al., 2007). Studies that focus on the short-term pre-disaster state indicate how women are disadvantaged by inadequate warning systems and cultural and economic choices on whether to evacuate. Flood preparation includes issuing warnings, building embankments and making provisions for evacuation, but official warnings are often not widely available, timely or reliable (Islam et al., 2012). Community knowledge is particularly important here as the best, and considered most trustworthy, warnings come from local experience and understanding (Manock et al., 2013). Initiatives that build on this research involve women becoming involved in early warning processes via door-to-door information campaigns, for example a recent project undertaken in Laos (Trohanis et al., 2011) helped to break down social as well as technological barriers to warning system access.
Pre-crisis literature can also take a long-term pre-disaster approach, focusing on broader socioeconomic and cultural factors that make women particularly vulnerable to disaster impacts. While this is linked to general vulnerability research, pre-crisis research tends to focus on ecological or environmental changes. In particular, in recent years there has been increasing study into the way climate change and environmental degradation impact upon women in Bangladesh and similar areas. Momtaz Jahan (Jahan, 2008) has identified a number of ways environmental degradation increases hardship and vulnerability to flooding in Bangladesh, particularly for women. According to Jahan (2008), increased salinity increases the risk of flash flooding deforestation, while deforestation and rapid urbanisation decreases vital resources and access to these resources (Jahan, 2008). For Jahan, women’s responsibility for the immediate survival needs of their family means that deforestation and urbanisation are particularly devastating (Jahan, 2008). The effects of climate change are forcing women to work harder to collect food, fuel and water (Baten and Khan, 2010). This is particularly concerning in developing nations, where people rely more on immediate natural resources for sustenance and rebuilding (Enarson, 2002). However, it is important to note that disasters become catastrophic because of political choices about people and land (Enarson, 2002) and so political and government choices can play a part in pre-crisis vulnerability.
During the crisis
Research into the crisis phase focuses on the immediate problems women face during floods and other disasters. Women are most vulnerable during the disaster phase as they take on extra burdens and have dramatically reduced support. Disasters increase women’s domestic and emotional labour while at the same time undermining their social protection (Enarson, 2002). Deeply embedded social roles mean women are responsible for fuel and water needs and during disasters have to wade through floodwaters for long distances, making them more prone to disease and being swept away (Rashid, 2002). Many people fear snakes, drowning and electrocution (Rashid, 2002). Because of this, as well as lack of work, many male family members leave the area to find work elsewhere (Ahmed et al., 2007). As they tend to own fewer valuable goods than the men in the family, women’s goods are usually sold first, leaving them worse off than before (Hanchett et al., 1998). The most comprehensive overview of this argument comes from Elaine Enarson (Enarson, 2002) who argues that disasters destroy land, robbing women of income and leading to poverty. She also claims that disasters increase women’s domestic and emotional labour, undermine women’s social protection and expand women’s voluntary work. Apart from a comprehensive description of socioeconomic disaster impact on women, Enarson makes two important points. Firstly, she notes that disasters bring to the fore survival needs of women and their families, meaning small women-operated enterprises that existed prior to the disaster are often abandoned. Secondly, she claims that the expansion of women’s voluntary work also means the expansion of their public work and that this can increase their status and awareness of their own capabilities, possibly leading to them being seen as potential earners. In particular, Enarson emphasises that simply restoring the status quo of women will leave women just as vulnerable as previously, meaning that efforts must be made to improve the pre-disaster status of women even when facing the practical challenges of post-disaster
Running alongside the extra socioeconomic burden is the social and cultural role women inhabit. This role narrows their options and increases their psychological burden during disasters. Social norms such as the restrictive movement of saris, or the way women are expected to stay inside homes, has often been attributed to the higher mortality rates of women in the crisis phase of disasters (Rashid, 2002). Gynaecological problems and sexual harassment are also reported to increase dramatically for women during this period (Ahmed et al., 2007). Toileting in flooding is particularly traumatic for women, as is menstruating, as these activities are heavily regulated by ideas of modesty and shame (Rashid, 2002). Crisis shelters are a major concern for women during flooding. Shelters allow for little privacy and so female bodily functions become hard to hide and even sleeping becomes difficult (Rashid, 2002). In fact, going to crisis shelters is often a source of shame in itself and many women choose not to go, though it is important to note that this is often a rational decision based on weighing up the minimal relief they would receive and the need to guard possessions (Khondker, 1996).
During disasters, and in fact all phases of disaster, social capital becomes particularly important. Social capital can be described as norms, networks and other factors that facilitate coordination and cooperation (Ali and Niehof, 2007). According to Ali and Niehof (2007: 244), social capital is a ‘crucial factor in averting vulnerability because it provides access to resources’. Women without social capital often sell their assets or incur heavy debts from moneylenders (Rashid, 2002) and kin or neighbours are generally approached for food gifts or loans before seeking loans from NGOs or rich people in the village (Ali and Niehof, 2007). On a community-wide level, it has been found that communities with more equal distributions of wealth in Bangladesh are more likely to spend resources on collective flood protection (Brouwer et al., 2007). Though it has been described as one among five different vulnerability components (Cannon, 2000), others have argued that it is one of the most important elements (Nakagawa and Shaw, 2004). Though disasters have both institutional and social aspects (Islam et al., 2012), research on disaster and gender in Bangladesh tends to emphasise the importance of social capital. This may be because institutional responses in Bangladesh are generally considered inadequate, with government largely considered corrupt and ineffectual and even NGOs are not always utilised by locals. It may also be because focusing on enhancing social capital provides a realisable and bottom-up approach to enhancing disaster resilience for women. Pallab Mozumder (Mozumder, 2008), for example, analysing health and social insurance after the 1998 flood in Bangladesh, noted that households with stronger social bonds spent less on medical expenses when recovering from floods.
Post-crisis
The post-crisis disaster period is key to breaking vulnerability patterns and improving future disaster resilience. If disasters exacerbate existing social structures and problems, simply restoring the status quo leaves women just as vulnerable (see, for example, Enarson, 2002) so gender must be actively engaged in rebuilding efforts in order to improve on this pre-disaster position. According to Islam et al. (2012), post-disaster response involves two phases: recovery and adaptation. Bangladesh has learnt lessons from previous disasters, shifting policy focus to flood proofing rather than just flood control, and improving access to important food sources (Beck, 2005), however without a gendered lens the improvements do not address gender inequality. Many theorists argue that in the recovery phase, disaster relief efforts actually heighten and entrench discrimination against women by excluding them from labour and decision-making. Though this is related to the way disasters exacerbate existing social structures, this argument is not focused so much on the pre-existing vulnerability of women to disaster impacts, but on the way the disaster events themselves change the expected behaviour and social roles of women. While many of these disaster changes are inextricably linked to prior social roles, it is important to note the subtle difference in focus, as this focus premises the disaster event itself as creating different forms of vulnerability. According to this idea, new types of gender discrimination can emerge from disasters which, although not independent of existing social structures, are a direct result of disaster (United Nations SDD, 2010). This is true even of relatively equal societies, such as the retreat into traditional gender roles in post-bushfire California (Hoffman, 1998).
Disaster relief and rebuilding programs may bypass women by excluding them from rebuilding work and the decision-making process. Women’s work is not considered important and post-disaster rebuilding efforts are male-centred, focusing on rebuilding infrastructure rather than livelihoods (Beck, 2005). Rebuilding efforts therefore, tend to employ men, rather than women. This means that the position women were placed under in the crisis phase is likely to continue. Of particular concern is the idea that women are excluded from decision-making processes in disaster recovery. Though a prevailing societal problem, it is argued that disasters exacerbate this and even do so in communities where gender differences are not usually so apparent. For example, Enarson and Morrow (1998) describe the way disaster recovery in Miami in 1992 led to the establishment of a women-only organisation Women Will Rebuild in response to the male-dominated relief effort We Will Rebuild. According to Enarson (2002), it is important to recognise how the organisational structures of disaster recovery are gendered and shaped by a male workforce and male culture. Despite the challenges of male-centred disaster recovery processes, research into rebuilding tends to focus on the capacities of women, rather than their vulnerability. It is worth noting briefly here the difference between framing disaster in terms of vulnerability and impact, and framing it in terms of response. While the two are closely linked, the first is passive and the other active.
In Bangladesh, disasters change work life and conditions, and gendered rebuilding efforts target women’s agricultural and land rights. Women are being recognised as more important in agricultural activities than previously thought and are in charge, for example, of storage and germination of seed grain (Hanchett et al., 1998). Similarly, many also argue that the focus of gendered disaster relief should be for women to regain their entitlements, such as ownership of poultry and cattle (Khondker, 1996). In a study of women-centred disaster recovery projects in East Asia and the Pacific, the World Bank recently documented a number of case studies in which preserving women’s land rights after disaster has become a priority, with recovery efforts focusing on rebuilding the land administration system and registering female, as well as male, heads of houses as land owners (Trohanis et al., 2011). After disasters male family members often move away to find work. This has led to a feminisation of agriculture, with women moving from unpaid farm workers to farm managers, though their involvement is still mostly with livestock and poultry, rather than crops (Jaim and Hossain, 2011). Fishing, one of the main industries in Bangladesh, has also been changing due to disasters. Women are participating more in fishing, as well as shrimp farming and snail collecting, though they are still restricted in the way they fish and do not make fishery management decisions (Sultana et al., 2002).
In response to this, disaster relief and long-term development projects focus on helping women to build their socioeconomic capacity and status. Often these projects are tied in with improving social capital. For example, Microcredit schemes put in place in Bangladesh foster social capital by building strong networks of trust and reciprocity (Mozumder, 2008). Banchte Shekha, an all-female NGO in Bangladesh, has formed women-only groups to which they provide fishing training, mobilise groups to save money, provide credit, have local meetings and form management committees to manage sustainable fishing (Sultana et al., 2002).
One would expect that the need for women to manage their own (and their family’s) affairs will become even more important in circumstances where men are absent, or absent for extended periods of time. This is increasingly the case in Bangladesh, however commentary on the gendered nature of disaster is scarce. The goal of our research is to establish how the absence of men impacts on the response of women within the cycle of disaster.
Research method
The district of Sirajgang in Bangladesh was chosen as the site of our research because we had been made aware that the villages within this district were extremely poor, particularly susceptible to flooding and there were few healthy adult males present for a significant part of the year. Consequently, it was expected that women would be assuming unfamiliar roles within family and village life. Within this district, the area of Shimla (consisting of two adjacent villages) was chosen for our case study. Each of the six (female) respondents were interviewed in depth individually on the basis of an open ended questionnaire and in the privacy of their own homes. The interview participants were between 19 and 29 years old, married with at least two children, and none had completed more than primary level education.
The interviews were conducted in the local language, taped, transcribed and translated into English. The respondents were interviewed by one of the principal researchers assisted by two female research assistants with good working experience in the field and who were capable of taking detailed notes of their observations. The interviews were all based on the same questionnaire which contained a number of open ended questions related to individual–societal aspects of the role of women in risk-management, division of labour, risk experience of women, gender norms and women specific strategies to cope with disasters, women’s leadership roles, discrimination against women, women’s social, economic and cultural vulnerability and their well-being, and women’s role in economic development. Follow-up questions were generated by the initial responses. The translated transcripts were subjected to content analysis to provide the findings reported below.
The translated transcripts were subjected to content analysis to provide the findings reported below. The transcripts underwent a process of data reduction with similar topics being coded so that similarly coded passages, by the same respondent and across respondents, could be considered at the same time. Once coded, responses underwent a process of abstraction in order to combine categories that appeared to belong together and of comparison in order to understand the similarities and differences occurring in the responses.
The verification of our data relied, to a large extent, on this iterative process. In addition, an element of triangulation was possible as each of the four researchers (from different research backgrounds) independently analysed the data. The reflections of the researchers, particularly the researcher who undertook the field work, were regarded as an invaluable source of data.
Findings from the Sirijgang case study
The reason for the absence of men
As previously noted, the villages of Shimla contained few men. The men are often away from the village for long periods because they are engaged in occupations for which there is little or no work in the village and they have to travel considerable distances to obtain work. Generally, working in the larger cities provides the only opportunity to gain consistent employment because, at best, only spasmodic employment can be obtained in the village. In our interviews we found that these men worked in a wide range of occupations, but almost invariably at the lower end of the income scale. One respondent’s husband was working in Dhaka as a bricklayer. She explained that her husband was unable to come back to the village on a regular basis because of the expense of travelling. As a result he was absent for more than half the year in order to take advantage of the job opportunities and better pay he experienced in Dhaka (Interview 1). This was not an uncommon experience. Another respondent’s husband had worked as a rickshaw puller in the village, but was able to get a much better paying job as a brick-maker in Dhaka (Interview 2). Although hard on the family, the opportunity to work in Dhaka cannot be turned down as it is often seen as the only means for the family’s survival (Interviews 1, 2). However, Dhaka was not the only alternative source of employment. One woman’s husband travelled over 200 kilometres to find work in the brick-fields of Takurgon in Dinajpur district. The work was seasonal, but meant that he would be away for up to six months working for 12 hours per day at a job that earns just above AUD$3 per day (Interview 3). Another husband became a hawker (mobile trader) and was forced to travel extensively throughout Bangladesh in order to make a living (Interview 4).
In one of the more entrepreneurial families, a husband had been able to establish an effective small business after acquiring a sugarcane crushing machine which enabled him to sell juice in a market 150 kilometres away. Again, this business required him to be absent for all but a few days a month for at least six months of the year (Interview 5). Apparently, by travelling away from his home village, he had been able to establish a monopoly in the city market and was able to make what was considered a good living. He had been prevented from doing this in his local village because there were too many competitors locally in this business. Had he stayed in the village he would have been forced to work as a labourer in the rice fields – if, indeed, any work could be found (Interview 6).
Clearly there were two advantages to the men moving away from home. The first was that it was their only reasonable hope of finding secure work. Secondly, the work they did find was generally better paid than any local work they did manage to pick up. By travelling to find work and by living away from home in order to work the jobs they found, the men were substantially adding to the meagre resources of the family. However, the pay they received was still so low that it was simply not possible for them to take their families with them as the accommodation costs would have been prohibitive (Interview 6).
Problems generated by the absence of men
The additional income that was generated by the men working away from the village was always welcome, however it clearly added to the difficulties confronted by the women. They found the most simple of living tasks difficult, if not impossible, to achieve in the absence of men because of the long standing cultural expectations of the community. Women were expected to work in the home and to rarely venture out. However, all were confronted by the need to do the shopping in the village, to take children and the infirm for basic health care visits to a doctor, and to ensure that the children were safely escorted to school. Although some of the women did, at times, endeavour to undertake these tasks, it was generally only under exceptional circumstances because of the cultural pressures placed on them (Interview 1).
A very unsatisfactory alternative, which actually placed more stress on the women, was the decision to allow young sons to undertake activities that would normally be the responsibility of the father (Interview 1). In one case, a 10-year-old boy was sent out to do the family shopping (Interview 2). All of the women expressed their concern at children being denied the benefit of a two parent family.
Further, although not a cultural issue, tasks at home requiring the strength of a man must be overlooked for months on end. Many of these tasks are essential prerequisites in preparing for flood. For example, the repairing of fences and the collecting of firewood (Interview 1), the moving of heavy items, such as large bags of rice, the building of higher platforms for storage (Interview 2) and undertaking repairs to the home (hut) which are too expensive for the women to have a labourer do (Interview 3). The long absence of men also makes it difficult for basic maintenance to get done (Interview 3) and as a result things deteriorate much faster (Interview 3).
An additional problem that women are often confronted with is the need to make decisions that would normally be taken by their husband. From our responses, it was clear that the women would prefer not to be placed in this position – again because it is culturally unacceptable (Interview 1). Although they were able to discuss their decisions with the men when they were at home, this is not the same as being able to call upon them when they need help and adds to their feelings of insecurity – especially when they are well aware that disaster could strike at any time (Interview 2).
Exacerbation of problems during floods
The experience of flooding during the absence of men is regarded as very painful by the women who are left to fend for themselves and their families (Interview 1). It is the women who must secure family belongings from flood damage. Food has to be placed in a protected area, along with firewood and cooking utensils. Any livestock must be moved to higher ground and protected from rising water (Interviews 1 and 4). As floods worsen, women are forced to take the family (children, elderly and disabled) to safer areas – again transgressing the cultural taboo of women operating outside of the home. Making matters worse is the fact that flood shelters are relatively few and quite crowded. Often, they are restricted to gathering on a relatively safe section of highway or river embankment with other families in similar positions (Interview 1). Under these circumstances providing for the family becomes even more difficult and people count themselves fortunate if, under these circumstances, they are able to get one meal a day (Interview 1). While such events would be trying under any circumstances, dealing with the problem while the husband is absent increases the level of stress experienced by these women significantly.
Even the first response to flood warnings is hampered by the absence of the husband. Flood warnings are provided, but communication is difficult when people do not own televisions or radios (Interview 3). Warnings are generally passed on by word of mouth, but if you are a woman and culturally expected to remain at home, being in a position where you can actually receive the warning in a timely way is problematic (Interview 3).
Social resources and social capital
All of the respondents reported that their husbands regularly forwarded a substantial portion of the money they were earning. This money is generally used to obtain every day necessities, for the schooling of children (Interview 2) and for the care of elderly relatives (Interview 3). However, the money is usually not enough to meet the basic needs of the family (Interview 1). The women feel guilty that their husbands have been working so hard to obtain the money and that they have not been able to stretch its use to meet all of their needs. Frequently, when there is a shortfall, women are forced to seek help from the local community in the form of loans from neighbours and the relatively better off members of the community (Interview 1). These loans can mount up quickly and much of the money the husband eventually brings home disappears in repayments. This becomes a cycle of living from which the women consider there is little chance of escape (Interview 1).
An alternative to borrowing from neighbours is to borrow through the Grameen Bank, which provides micro-credit loans in rural Bangladesh (Interview 5). Although this is a more formally institutionalised form of lending, the financial stress it places on the women can be very high.
Making matters worse is the fact that women, in an effort to avoid debt, often go without adequate food in the absence of their husband. They conserve their resources by cooking just one meal a day, although when the husband returns home he is provided with three cooked meals a day (Interview 3).
Without the help of the local community, it is doubtful that many of these women could survive. Community assistance during flood events is expected and considered very much a part of the way of life in the village community (Interview 1).
Investing in the future
Coping with disaster when it occurs is, by itself, not enough. These women are clearly aware of the need to provide for the future and regularly seek ways of increasing the family’s resources – even when they know that their chances of success are limited. For example, one woman had planned to increase the family’s holding of livestock by rearing two calves. However, the flood waters had polluted the fields and the calves died from unknown diseases. ‘Allah has taken these resources away from me’ (Interview 1).
However, her failure has not deterred others from trying and, at times, being quite successful. One respondent told us:
I received two small cows from an NGO about two years ago. When they were grown up, and delivered calves, I sold them for good money. The proceeds enabled us to buy a small piece of land for cultivation. Rice is the only crop we get from this piece of land. The crops provide us rice which lasts for about three months. My father in-law and brother in-law help us to look after the crops, cut the crops and bring the crops home. (Interview 2)
This same woman (Interview 2) was able to do simple sewing tasks to supplement the family income. Other means adopted by the women to try and earn extra income include the local purchase of vegetables which are then taken to nearby towns to, hopefully, be sold at a profit. They also endeavour to grow their own vegetables for sale, but this practice is hindered by the fact that they generally have to rent land and pay the landlord half of the crop. This latter practice, known as barga, is a traditional custom (Interview 3). If there was spare money available, some of the women would also like to go to the city to purchase goods which they could then, hopefully, sell at a profit in their local area (Interview 3).
Although it would appear that these women are often operating independently, this is not always the case. In one instance (Interview 3), practices designed to earn extra income had to get the approval of the husband. As poor as these people were, this often meant keeping in contact with the husband via mobile phone:
I have a mobile phone. I speak to my husband quite often, he gives me instructions, suggestions. My husband is quite strong in his personality, so I discuss every plan with him first for his approval. (Interview 3)
Indeed, the use of mobile phone technology was quite surprising. It enabled couples to keep in contact (and the male to hold control over decision making) and it was also through the mobile phone service that money transfers were frequently made (Interview 3).
However, the women are also clear about some of the things that hold them back. The woman referred to above was planning on building her cattle herd, but considered that her lack of knowledge with respect to breeding and feeding was a significant handicap which was made worse in times of flood, while another woman mentioned the high cost of maintaining the cattle (Interview 2).
The women are quite clear about what they would do if they were ever to have an increase in the resources at their disposal. They would act to provide themselves greater protection from flooding events. ‘I will re-build my hut/house on a higher base/platform’ (Interview 1). However, rebuilding often means taking what they can salvage from the wreckage of their old home and utilising it to construct a new dwelling or temporary dwelling (Interview 4). Further, they recognise that this will only protect them from lower level flooding events (Interview 1). Others would like to use any government funding to purchase assets that can be used to raise income in the future (cattle for example) and one of their most cherished ambitions is to purchase land on a higher level and which therefore has some protection from flooding (Interview 2).
One respondent once owned their own home, but this was destroyed by flood and resulted in them having to move in with their in-laws. This woman considered that she was still relatively well off as they were not paying rent as was the case for many others (Interview 2). However, she recognised the very real possibility that even this home could be lost in the next flood.
Some people are careful not to purchase land near the river, but to try and find small parcels of land on higher ground which they considered a more sustainable investment (Interview 3). But working the land is problematic. Some women have fathers or father-in-laws who assist in the absence of the husband, but often the women are forced to hire labour which they cannot really afford (Interview 4). In a good season the work that women do in growing food, in particular rice, can add substantially to their income and often growing rice is considered more sustainable than having animals as they do not have to worry about trying to move them to higher land in time of flood – an activity that is hampered by both its physical requirements and the higher priority of looking after children and the infirm (Interview 4).
Building resilience
The experience of coping with flood without the assistance of the husband is not new for these women. They have gone through floods many times (Interview 1). As a result, the women frequently meet together to discuss how they can be better prepared to meet the problems they know they will experience during the flood. Sometimes these meetings are coordinated by a local NGO. One of the outcomes of this activity is that women become more confident that they can handle a flooding event when it occurs. The local NGOs also endeavour to provide advice on how to handle flooding events (Interviews 1, 2). However the government plays virtually no role here: ‘They don’t come to see our misfortunes; they don’t really discuss anything with us. We are on our own’ (Interview 1). While the women seek to protect themselves from lower level floods, they recognise that they need government help to protect them from more severe flooding – ‘government better allocate money to repair and built river embankments to protect us from this recurrent floods every year’ (Interviews 1, 2, 5). Indeed, having the government undertake this type of work was a common desire amongst the women (Interview 2). All of the women wanted to see better river embankments built (Interviews 3, 4). Occasionally the local council will provide a small quantity of rice, which by no means sufficient, does at least assist (Interviews 3, 4).
The women are well aware that they are pretty much on their own in caring for their family and securing their future. They are conscious of their need to continue to improve their position and skills in order to enable them to do this. ‘More and more I am gaining the skills which is needed to manage the everyday running of my family without the help of my husband’ (Interview 2). However, as they are forced to live in somewhat remote locations, the ability to access training is virtually non-existent and they are almost totally reliant on NGOs bringing the training to them (Interview 3). Much of the training is in the handling and feeding of livestock (Interview 6). They also recognise that they cannot let all jobs wait until their husbands are able to return home:
While my husband is away, I have arranged to put the fence at the front of my hut. I have saved a little bit of money to buy few more corrugated sheets for the fence for the back of the house. My husband doesn’t know carpentry well. I don’t think I can wait for him to take care of everything. He works very hard in the brick field, he needs a bit break, and I cannot pile up unfinished jobs for him. (Interview 2)
Although these women see themselves as living a ‘hand to mouth’ existence, and although they would like to be able to save, they know that there may be no practical opportunity for them to do so (Interviews 1, 2). To them, the best protection they can take is to keep ready, to keep in contact with their neighbours, to take the advice of the elderly (who have been through it all before) and be ready to move when the need arises (Interview 1). They are also very confident in their ability to use the resources that are at their disposal in the time of floods:
But most people in this area have dinghy boats, we borrow from them if we need to relocate heavy things (for example, bed, furniture etc.) to a secure highland. I can operate a dinghy boat myself, so in emergency I feel quite prepared. (Interview 3) We are well trained operating our own dinghies, we born in water, live in water. It’s our lives. (Interview 1)
They consider that NGOs are typically very helpful and provide advice on what to do before and after a flood and even help provide them with limited resources – such as a cow (Interview 2). Often the information provided is extremely basic – like keeping some food dry – and one wonders why it takes an NGO to provide this type of information. Why, for example, is it not part of the school curriculum? Apparently for cultural reasons, the NGO field officer that visits is a woman (Interview 2).
A sense of community is essential given the circumstances under which they live:
During the flood we arrange to remove children, elderly people as our priority. Also, if someone is pregnant and in her advanced stage, she gets the priority. Everyone understands this priority, I don’t think we have to teach anyone these practical things. These are not individual issues, these are community issues and we work as cooperatively as possible. Neighbours, who own house-boats, always lend them to others to relocate these vulnerable people first. (Interview 2) We don’t have any boat, and I cannot operate a boat. Anyway, my neighbours help me in the process of preparing for a flood and if relocation is needed I have no problem relying on other people in the community. (Interview 4)
Whilst many of the women in the area were looking for ways of getting out, others were moving in. These people needed to be instructed in what to do in time of flood and this was frequently left to the women to do:
I talk to people who are new here and advise them what should be done to cope with the flood. This is an everyday matter for us, some of us have more experience than others, that’s all. (Interview 3)
This will include older women instructing younger couples how to respond to a flood situation (Interview 4); a sense of inevitability pervaded the responses. Many were not living on their own property and had often lost property to the floods. However, this did not seem to bother them as their lives were determined by Allah (Interview 3).
Conclusion
Two million people were affected in the floodplains and low-lying areas of 56 unions of Sirajgang in 2012. Five villages surrounding the area known as ‘Shimla’ were lost due to river erosion when several flood protection embankments collapsed. Seven hundred and fifty families were made homeless and forced to live in small temporary huts on the surviving river protection embankments. In these areas, unemployment rose alarmingly and the jobless left their villages to find work in larger cities, leaving behind their vulnerable and insecure families. As a result of these drastically changed circumstances, women are increasingly required to take on totally unfamiliar roles, having lived their entire lives in a traditional Islamic society. The changing demands on these women poses an immediate threat to long established social structures.
The research in Shimla utilised in-depth interviews with women managing without the support of their husbands. We found a complex interplay of individual and societal aspects related to the risk management and coping strategies of the women. The analysis sheds light on the leadership role of women, their social, economic and cultural vulnerability, their psychological recovery and their struggle to maintain family well-being.
Our research suggests that the issues faced by women during disaster are compounded by the absence of their male counterparts throughout many of the disaster phases. In addition, our research establishes that the level of resilience amongst these disaster-affected women is not easily explained, especially when considering their low economic and social standing within the local community. Their stoic endurance of all that is put in their path indicates a resource which the authorities in disaster prone countries such as Bangladesh could utilise. Engaging with the poorest flood-affected women to improve community resilience would involve increasing institutional capital, in order to both complement the effort that the women are putting in on their own behalf, and in order to enable these courageous, industrious and community minded women to leverage off, and substantially expand, the resilience of their communities.
Footnotes
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The field expedition to Bangladesh in December, 2012 was funded by the Faculty of Arts, Charles Sturt University, Australia.