1. Bangladesh, a flat low-lying riverine country is located in northeast South Asia and is part of the humid tropics. Its 600 kilometer-long coastline touches the Bay of Bengal in the South. The country's peculiar geography and climatic conditions make it one of the most disaster-prone locales in the world. The most common disasters are floods, riverbank erosions, tornadoes, thunderstorms, tropical cyclones and the accompanying surges that form in the Bay. But the most treacherous have been cyclones, responsible for the largest number of disaster-related deaths, particularly in villages in the coastal districts of South Bangladesh. The cyclones of 1970 and 1991 claimed 500,000 and 138,000 human lives respectively, in addition to the destruction of other life forms and infrastructure. The largest numbers of casualties and deaths were those of women, children and the aged. More than half of the 138,000 deaths in 1991 were reportedly from this more vulnerable population groups (Haidar, Rahman, and Haq. 1991).
2. Bangladesh is also one of the largest of the world's least developed countries. It had a Gross National Product per capita of only US$ 235 in 1996. Its Human Development Index at 0.371 fell among the least developed countries, which had an average HDI of 0.344. The Gender Development Index of 0.342, compared to 0.564 GDI for all developing countries (UNDP, Bangladesh, 1999). Bangladesh has one of the highest population densities (800 per sq km) in the world, an adverse population resource ratio and a slow pace of economic growth. Despite recent attempts at industrialization and poverty reduction measures, Bangladesh remains a predominantly agrarian economy with agriculture contributing 30 per cent to total GDP and 60 per cent to total employment (Mahbub ul Haq. 1997). An absolute number of 43 million people live below the poverty line of 2,122 Kcal. Nearly 60 per cent of households own less than 0.50 acres of land and 70 per cent own no productive assets. High levels of illiteracy, malnutrition, morbidity, mortality are related development problems.4
3. The southern coastal villages are by and large endowed with rich natural resources: fertile alluvial soil, rainfall and marine resources. This, coupled with the opportunity for small capital investments in certain economic activities, heightens the income-generating potential of these areas. However frequent cyclones and flash floods have a devastating impact on all life forms, property and infrastructure; exacerbate class, ethnic and gender disparities; and make long term recovery, especially for disadvantaged segments, of the population more difficult.
4. The dominant economic activities in these areas are farming and fishing. The main crops produced are paddy and wheat under the share cropping system. Salt and betel nut production, artisanship, trading, petty vending, and employment in services as village teachers, as village-based NGO workers or in the town-based service sector are other important occupations.
5. A low level of infrastructure development marks the villages. Most houses are made of straw and bamboo and suffer severe damage during disasters. Villages have cyclone shelters, but lack proper, well-constructed approach roads especially to the shelter. This coupled with the inadequacy of shelters and their distance from homes makes access to shelters difficult. Apart from the elite few who use deep tube wells, most of the population uses shallow tube wells. The water is unclean and in the dry season water levels fall, making access to water difficult. Most villages lack sanitary latrines, electricity, accessible medical services and well-constructed poultry sheds and killas, specially prepared high ground for the protection of livestock during cyclones and floods.
6. Ownership and control over productive resources, income and wealth is skewed. The apex of the rural economic structure is dominated by agricultural landowners, salt pan owners, owners of fishing trawlers and hatcheries, moneylenders and bigger traders, while agricultural laborers, tenant farmers, salt pan workers, workers employed on large fishing trawlers and hatchery workers form the base of this hierarchy. Between these extremes are small peasant farmers, petty traders, and smaller independent fishermen. Activities are seasonal. Livelihood sources and resource ownership is diverse. An agricultural landowner may also own salt pans and hatcheries; and salt pan workers who engage in salt pan production during the dry season from October to February, may also hire themselves out as agricultural laborers during the farming season, and engage in some petty trade throughout the year.
7. Another component of the socio-economic hierarchy are the religious leaders, who not only own and control economic assets, but are the interpreters and keepers of the community's religious conscience.
8. The experience of impoverishment and deprivation in Bangladesh is more acute for women. Economic impoverishment interacts with male-oriented family and kinship, community, market and state structures and processes, marginalizing women more than men, economically, politically and socially.5
9. While patriarchal contexts concentrate power and authority in male hands both within and beyond the household, discriminatory patrilineal family and kinship systems accord girl children, unlike boy children, partial membership in natal families. Girls are transferred on marriage - their socially prescribed destiny - to their marital homes, where they are supposed to belong. They do not ensure the continuity of lineage, nor are they supposed to provide support of any kind to their natal families after marriage, this being perceived as a slur on family prestige. This results in a widespread culture of son preference that leads to a greater investment of material and non-material resources in sons than in daughters. This coupled with gendered constructions of femininity - gender role and trait stereotypes - circumscribes a woman's activities and possibilities, and marginalizes her from ownership, control over and access to material and non-material resources both within the household and in society at large. It is no historical accident, therefore, that in the coastal villages, (as in other parts of Bangladesh) ownership, control over and access to productive material resources like land, salt pans, hatcheries, credit, equipment and the produce, and material benefits accruing from economic activities related to these resources, rested with men.
10. Principles of patrilineal descent place considerable emphasis on the biological paternity of a child result in a stringent control over a woman's sexuality and reproduction at every stage of her life. Bodily integrity and sexual purity are important features of virtuous womanhood, and determinants of familial and community honor. Men are entrusted with the task of guarding and protecting this virtue.
11. There have, however, been some transformations in the last couple of decades that have impacted different categories of women in Bangladesh differently. New economic opportunities have emerged for middle class urban women, drawing them into paid public employment. The growing landlessness and impoverishment of the majority of the rural population, and the consequent erosion of the basis of their productive role within household-based production, is increasingly compelling poor rural women's entry into the labor market. However restrictions on women's physical mobility have contributed to confining them to the informal, undervalued and invisible margins of the labor market. In general their paid work has represented an extension of their traditional roles. Employment in the export oriented garment industry, has been one option for poor women since the 1980s in urban areas. Domestic service, begging and prostitution are other alternatives.
12. It is this socio-economic, political and cultural landscape that frames women's vulnerabilities and capacities in and in the aftermath of cyclonic disasters.
13. The vulnerabilities and capacities of men and women in the cyclonic disasters of April 1991 and May 1997 are presented in terms of key issues and trends that emerge from the accounts of women and men in the disaster-affected villages and that of other stakeholders. They traverse various phases of the disaster cycle: the immediate pre-disaster phase; during disaster that includes evacuation and emergency shelter; and the post disaster phase that includes relief, rehabilitation, recovery and resettlement. The findings are not conclusive, but suggestive. They uncover issues for further investigation, important among which is an impact assessment establishing the interaction between gender and other social categories like class, religion etc.
14. In the cyclone of 1991, large numbers of women within the home or homestead died because warning signals did not reach them. In a highly sex-segregated society, warning information was transmitted through microphones, megaphones and radios (for those who possessed them), by men CPP volunteers to men in public spaces - main roads, markets, tea stalls - where men congregated, on the assumption that this would be communicated to the rest of the family. This by and large did not occur. For many women who did hear it, the warnings were not clearly audible, because of the wind sound and speed. Even when women clearly heard the warnings, they were ill-informed about cyclones, the nature and significance of warning signals, the kind of household and community-based measures to be taken at each signal; and the need for timely evacuation. They were dependent on male decision-making about the course of action to be adopted. Like the men, most ignored timely warnings believing that nothing would really happen because the previous cyclone of devastating proportions in 1970 had occurred nearly twenty-one years earlier and was therefore faint in collective memory. Many had never lived through a cyclone like this. As a result only twenty-five per cent of the population in badly affected coastal areas evacuated (Haidar, Rahman and Haq. 1991). Moreover, being a sex-segregated society, women did not move even when warned, fearful of being labelled bepurdah (loose) and incurring the wrath of their husbands and the community, if they ventured out, unescorted by male relatives. The fear of sexual harassment and the fury of the weather in the dark of night were other factors that compounded women's insecurity, immobilising them. In the ensuing procrastination, many women perished with their children while waiting for their husbands to return home and take them to safety. Others decided to evacuate too late. Still others who in their socially prescribed domestic roles bore the lone burden of trying to get together children, the aged, their livestock, poultry and other belongings to evacuate, were delayed and perished in the surge.
15. Some of the 1991 experiences were repeated in May 1997, especially where no disaster preparedness NGO activity had yet been initiated. These included late warnings, women's receipt of warnings later than men and a consequent lack of time for evacuation.
16. By contrast, in areas with community-based disaster preparedness NGO activity, women and men demonstrated better awareness and knowledge about the nature of cyclones, their impact and the nature and significance of warning information, and what household and community-based measures were to be taken at each signal. Warnings received over the radio, microphones and megaphones, as also by watching for hoisted flags on shelters, were timely. Primarily male community based disaster preparedness committee members and male Cyclone Preparedness Programme Volunteers, of Government of Bangladesh publicly communicated this information. However in certain areas women committee members were part of the community-based pre disaster emergency meeting, in which responsibilities for disaster management and response were allocated to area-based groups. In other areas women committee members, other local women and women Red Crescent Youth volunteers disseminated warning information by word of mouth to neighbors and friends within the smaller range of the neighborhood, while in still others women committee members even disseminated warnings publicly, using microphones and megaphones.
17. Women commenced their household-based preparedness activity in time. These included: safe storage and preservation of food, and valuable items in earthen pots underground, unfastening livestock and sending them with male family members or older children to be kept at heights on embankments; fastening houses to trees, and keeping ready emergency food, water and valuables to be taken to shelters. This contributed significantly to the preservation of life, especially those of women and children. Moreover women experienced a heightened sense of involvement in socially useful community activity.
18. A pervasive concern of women was the burden they bore of handling the entire responsibility of household preparedness measures. The men were either away working, in the vicinity chatting and gathering information on the latest developments, or in some cases herding cattle to embankments. Women expressed a desire for the men to be around as far as possible, and assist them with preparedness tasks.
19. In the cyclone of 1991, more women than men died while trying to save themselves and their children, due to the close bonding developed as a result of their mothering roles. By contrast, documented evidence points to men abandoning their families and saving themselves (Haidar et. al. 1991). A businessman from Boroghape going home to bring his family to safety, encountered rising water levels and decided not continue further. His wife and children died waiting for him to return. More men lived to narrate how their children had been swept away. In one instance, the father had to choose between holding on to one of his two children, as it was impossible to save both. He tellingly let go of the female child, and is reported to have said, "this son has to carry on the family line" (Ibid). Few women survived after they let go of their children.
20. In both cyclones, women found it more difficult than men to scale trees, beams of houses, rooftops, row boats and swim. In 1991, many perished in the surge with their children. This was compounded by their not being used to such vigorous `masculine' activity. For women are socialized to refrain from such activity in routine life, in conformity to prescribed social codes of modest and genteel feminine conduct. In 1991 in particular, this reduced their survival chances. In a culture that highly values female modesty, the dress code - the sari, the burkha and long hair, became a death trap for women, inhibiting quick movement and their ability to swim against the surge. In 1991, some women remained in the waters as they were disrobed in the surge and were ashamed to come out. Some perished, as a result. By contrast, men who were completely unclad unabashedly searched for pieces of cloth and even unrobed corpses to clothe themselves. This is indicative of differing male-female perceptions to body, sexuality and notions of modesty, which was again detrimental to women's chances for survival.
21. By contrast in 1997, in areas with enhanced consciousness women moved to shelters at signals 7, 8 or 9 in 1997. However, some families did not move, either because they were not part of the community-based disaster preparedness network and had consequently not yet recognized the need for timely preparedness initiatives; or they lived too far away from shelters; or they remained at home to pre-empt theft of their belongings and usurpation of their property. Most women had an agreement with their husbands to flee to shelters with their children, even in the absence of husbands or male family members. Women thus moved to shelters with their children, without fearing the wrath of absent husbands, or the community. Women maintained that they evacuated first with their children and the aged. If the men were at home, they either stayed behind to guard homes and belongings, went with the women or followed carrying the disabled.
22. Poultry and livestock were unfastened and generally left to fend for themselves, as there were a lack of appropriate killas.
23. Women who reached shelters in the 1991 and 1997 cyclones, found them too far and too few. Accessing shelters was difficult because of flooding, the consequent disappearance of approach paths, and uprooted trees that blocked movement. These conditions made it particularly difficult for women to, often single-handedly, move children, the aged and their belongings to safety. Women, children and the aged kept slipping and falling as a result of the heavy rain, high wind speed and slush. They were bruised from falling, from flying corrugated tin sheets and uprooted tree branches. Children kept getting lost. The situation was even more difficult for pregnant women. Women complained of sexual harassment and chain snatching en route to shelters.
24. Women faced specific problems in shelters. Both in 1991 and in 1997, they found them overcrowded and congested. According to the communities, shelters can on an average accommodate around 800 people, but during a cyclone the number is anywhere between 1500-2000. The shelters had no proper lighting facilities or arrangements for sex segregation. Women complained that the large numbers of men and women huddled together - a rarity in a culture of seclusion - heightened their insecurity.
25. Innovative attempts at sex segregation were attempted in some shelters - men being accommodated on the ground floor and in shelter verandahs, while women and children occupied the room on the first floor; or constructing makeshift partitions with benches that served the purpose of segregation. Shelters were not designed with separate toilets. There was no stored water or provision for toiletries like cloth for menstruating women and girls. Toilets in most shelters were constructed in the compound below, making it impossible for women and children to edge their way through the crush or access them in the midst of high wind speeds and pouring rain. Where shelters have a toilet within, they were not usable, as they had no water, or water connections, etc. Women and children therefore defecated and urinated in shelters, making conditions very unhygienic. All this reduced women's privacy levels, and especially enhanced the discomfort of menstruating, pregnant and lactating women. Further, shelters had no proper drainage facilities and rainwater entering through openings in the doors and windows flooded them. Women and older female children stood in ankle or shin deep water, carrying infants and little children to keep them above water levels. They suffered cramps and aches as a result. There were no ledges for women to keep their belongings, adding to the weight they were burdened by. Shelters had no proper provisions for adequate drinking water and lacked emergency medical and food supplies like puffed rice or high protein biscuits, which made it very stressful for the children and in turn for women with children.
26. In cyclonic disasters crops, vegetation, stored grain were either washed away or rotted. Salt ready for sale, hatcheries, fishing boats and nets were destroyed. Poultry and livestock perished. Standing saline water in the fields was a setback to cultivation. Rotting leaves and decomposed carcasses polluted drinking water sources. While entire families and communities lost their sources of livelihood and sustenance, the class and gender impacts of this economic loss were unevenly distributed. Those who lived in concrete structures with granaries adequately stocked, and assets carefully kept at home or in financial institutions, recovered faster. Small farmers, tenants and landless households suffered acute losses. Women from these households and women headed households lost life sustaining sources of nutrition and livelihood - food, food grains, seeds, crops, kitchen gardens and their products, poultry and livestock, stored water for drinking and other domestic purposes, jewellery and carefully saved income stashed away at home were even worse off. This negatively impacted nutrition levels of women and children; caused outbreaks of diarrhoea, dysentery, fevers, and respiratory and skin ailments; and resulted in women's greater exposure and susceptibility to some of these ailments in the course of nursing family members back to health.
27. In their nurturing and care-giving roles, and as producers, providers, and consumers of food and other goods and services for their families, communities and themselves, women suffered a sharp increase in work loads. In both cyclones, not only were women engaged in providing for the physical needs of the family - food, clothing, shelter, fuel, water, health care - but did so in difficult conditions and were encumbered by emergency operations such as the construction of make-shift shelters, constructing rafts and scaffolds to keep above water levels, sheltering animals, protecting their children and animals from insect and snake bites, taking special care of infants and the aged, particularly when ill. Women coped by invoking assistance from the village elite, borrowing from neighbours, scouring the environment for scraps of food and fuel wood, subsisting on emergency items stored underground or in the homes of the elite, or by going to live with relatives in less affected areas till the worst was over. According to both men and women, men were either away seeking relief items, opportunities to earn some income, or at familiar sites of male bonding like the tea shop, the local grocery kiosk or the main road chatting and exchanging information. Male out migration for jobs at a later stage was also reported.
28. Immediate relief and long term recovery support for income generating and housing reconstruction activities were distributed to men. Women in all these contexts report that male heads of households at times used relief items to suit their own needs and priorities, rather than those of the household (e.g. men spending money to buy cigarettes, tea, `paan' etc.). Despite their increased workload and creative contribution to the survival process, women, especially women heading households, were marginalized from access to the very relief items they were responsible for providing the family with in the post-disaster phase. This is because it was assumed that males head households and that there exists an intra-family equality (equality within the family) sharing of resources and harmony of interests. Agricultural and housing reconstruction inputs allocated by government and NGOs during rehabilitation are often tied to previous patterns of land and housing ownership, i.e. allocated to males who are the owners and perceived as farmers, or in the case of some female headed households these allocations were instead made to sons (even minors) or brothers of the late husband, instead of the wife. As women generally do not own land and houses, and are not considered farmers, they are once again marginalized from acquisition of such assets. Such marginalization coupled with the loss of their meagre resources in disaster exacerbates their impoverishment and slows down the pace of women's long term recovery even more than for men.
29. Women were disadvantaged in battling with physically stronger men in relief distribution queues and were hesitant to approach male relief workers in a sex-segregated culture. Carrying weighty packages home, the inundation of entire areas and disruption of normal communication systems, and observance of purdah were other factors contributing to this marginalization.
30. Relief food items and clothing were often inappropriate. Women complained that the distribution of dry grain in 1991, when it was difficult to find enough dry firewood to cook it, resulted in consumption of partially cooked grain and consequent stomach disorders, especially for children. Others said they sold the grain and children went hungry.
31. Men and women maintained that even in the cyclone of 1997, they had to wait for first aid and treatment for fevers and diarrhoea for themselves and their families up to three days after the cyclone. Doctors providing health services were men, thus inhibiting women's access to them. There was no provision for the psychological rehabilitation of communities both in 1991 and in 1997. Women are more emotionally attached to their children because they spend more time with them at home. The suffering and loss of children during the cyclones, the loss of male family members in a culture in which women are heavily dependent on men, the pressures of economic and cultural dislocation and the pressure on women especially to stabilize and normalize life for the family as fast as possible, makes the need for addressing their psychological needs (and that of the whole community) in disaster extremely important.
32. Finally, women who did receive housing reconstruction inputs after the 1997 cyclone, said that at times they were given only corrugated tin sheets for roofs and had none of the other related inputs for proper reconstruction. Also there was no assistance or training imparted to women to reconstruct. Even when female headed households, women from landless households or pregnant and lactating women were identified as priority target groups by the government and NGOs engaged in post disaster work, it was the immediate practical rather than the long term strategic gender needs of these groups that were addressed.
33. Emerging wisdom in disaster management highlights the importance of looking at people not just vulnerable victims of disasters, but as survivors who can contribute actively to the recovery process. This is especially true for women. In their socially constructed roles as nurturers, socializing agents and as key consumers of environmental resources, their capacities, coping mechanisms, survival strategies, local knowledge, leadership potential, skills and resources can be brought to bear significantly on disaster preparedness, relief, management and mitigation activities. Before the cyclone of May 1997, women with the help of the men replaced the old bamboo poles of their homes, raised the ground levels of their houses by adding more soil, built small `killas' for their cattle and poultry. Just before the cyclone struck, women placed essential household items, including food and clean drinking water in large earthen pots underground, as a safety and storage measure; packed emergency food items like puffed rice, water, and valuables like jewellery and documents in plastic bags to carry with them to shelters; tied household furniture to strong trees; collected and stored firewood; untied cattle and poultry either allowing them to move about freely or tying them to posts erected on `killas.' In the throes of and in the aftermath of April 1991 cyclone, women and their children drank coconut water till the first relief came. The survivors made make-shift bamboo and wooden rafts to float on and held on to logs and branches to keep above water level. They scaled and sat on tree tops and on the roofs of two storeyed buildings with surviving children. Despite this women's capacities remain even more concealed than their vulnerabilities in disaster.
34. Quite obviously the subordinate position of women in society enhances women's vulnerabilities in disaster and slows down their long-term recovery. Women's contribution to the survival process is perceived as a simple extension of domestic chores that is consistent with women's basic nature, and thus remains invisible.
35. Disaster management theory and practice contributes to this process. Disaster work has traditionally been a male terrain and largely continues to be so, because it involves public activity in difficult, dangerous, physically and emotionally draining conditions that are deemed naturally appropriate only for men. Its theoretical paradigms and practical interventions that are by and large constructed, planned and executed by men with gendered perspectives, consequently marginalize women's interests.
36. Mainstream conventional formulations on natural disasters defined disasters within the natural science paradigm as natural, exceptional natural calamities. (Hewitt, 1983b, pp. 5-7)
37. The 1970s and 1980s however saw a paradigm shift that framed disasters as a function of vulnerabilities of communities and an integral part of normal time development processes (O'Keefe et. al., 1976; Cuny, 1983; Maskrey, 1989; Cannon, 1994).
38. Both traditions, the natural science paradigm and contemporary vulnerability analysis when applied to communities as homogenous entities, are generic. They subsume women and their concerns under the rubric of a universal humanity. They disregard the different socially-determined role and trait stereotypes for men and women. They thereby assume a similarity of needs and interests between women and men; a unitary, harmonious, nuclear model of household, marked by a man as benevolent head and breadwinner and women in dependent domestic roles. They assume that women's class positions are neatly derived of from those of their husband's. They believe in an equality and mutuality of interests in patterns of resource ownership, allocation, control and use between genders within households and kinship groups, both in normal times and in disasters.
39. Generically oriented gender blind interventions thus conceal the special biologically and socially determined needs and capacities of women in disasters in relation to men, arising from their socially prescribed reproductive roles. They mask the differential impact of disasters on men and women, with women bearing the brunt. In effect they reinforce the greater marginalization and the burden women experience in disasters.
40. Following the United Nations (UN Decade) for women 1975-1985, that made it difficult for development theory and practice to marginalize women's distinctive experiences and concerns; the appallingly larger number of disaster casualties and deaths among women than men; the vulnerability approach's designation of women as a vulnerable population group and the insistence of international NGOs and donor agencies on addressing women as `relief beneficiaries' in disasters, women's concerns have now begun to figure on the agenda of disaster organizations.
41. Though this represents a step forward from the gender blind approach, these women specific interventions and their particular treatment of women's concerns leaves much to be desired. Women tend to be constructed as `biologically weak, inferior and hapless victims' with special biologically determined needs, needing to be protected and rescued by `stronger' men. This is attributable to the deeply entrenched male-oriented world view, culture and practices of these organizations that perceive male and female role and trait stereotypes and unequal gender relations, as natural and biologically determined. It is also perhaps attributable to the `historic links of these organizations to military and para-military emergency response' (Enarson, 1997), which is deeply male-centered.
42. This biologically rooted orientation of disaster organizations to women:
43. In assuming that unequal gender relationships are natural givens and hence unchangeable, this perspective lacks the methodological tools to move to a further level of systematic identification and analysis of women's concerns in disaster, in complex, changing social contexts. Women thus figure as mere additives and appendages on the disaster response agenda. It is therefore imperative to weave `gender' as an analytical category and `gender analysis' as a methodological tool of analysis into the core of disaster preparedness and management theory and practice.
44. Some disaster management organizations have however introduced innovative community-based gender initiatives that can inspire new thinking, adaptation and application in other contexts. Examples are those introduced by the Cyclone Preparedness Programme of the Government of Bangladesh, and the Community-based Disaster Preparedness Programme (CBDPP) of the Bangladesh Red Crescent Society (BDRCS). Both have been actively recruiting female volunteers and female field workers in their disaster preparedness and management work and providing gender training for staff and disaster affected communities. Further the BDRCS has adopted a village level community-based approach to disaster preparedness and management. This includes the formation of micro groups of men and women based on household clusters, and the recruitment of women and men field assistants. The programme has a clear gender perspective that looks at the differential impact of disasters on men and women, and seeks to rectify imbalances by addressing and engaging with both men and women. It involves both men and women in decision-making on disaster issues, and draws on the female assistant and micro group leaders to work with and train women in the community on household and community based preparedness measures.
45. There have been notable gender initiatives in other parts of South Asia as well. After the devastating floods of 1989 that destroyed the villages of Bharat and Shamsabad in the district of Muzaffargarh in Pakistan, the NGO Pattan catalyzed two important gender sensitive interventions in its reconstruction assistance. One was to incorporate women's suggestions into housing layout and design, the other was to introduce co-ownership of houses by husband and wife and involve the community in the decision making process of this project (Bari, 1996).
46. A similar project of joint ownership of housing by husband and wife, catalysed and funded by the World Bank, was implemented in post-earthquake reconstruction in Latur, India.
47. In highly sex-segregated societies in many parts of South Asia, for instance, the induction of female staff working closely with the community of local women and notions of co-ownership and shared responsibility of men and women help engender and enrich the practice of disaster organizations. More significantly this unleashes a new social dynamics of women operating in the public sphere and crossing their traditional boundaries, some becoming role models for other women in the community - a possible first step towards more empowering gender relations.
48. Moreover such initiatives catalyzed by community-based NGOs, existing local women's organizations or external organizations assume special significance in two contexts:
49. The foregoing discussion points to an urgency for a shift in the approach of disaster preparedness and management, so as to reflect the centrality of women's experience in pre, during, and post disaster situations, compare these with men and then embark on addressing imbalances. This will ensure women's rights as human rights, guarantee efficiency, promote family welfare and facilitate women's empowerment, thus promoting sustainable development.
50. To this end changes are specifically warranted on at least the five following counts: conceptual, legal, social, institutional and infrastructural.9
51. It is imperative to challenge conventional formulations of gender role and trait stereotypes and inequity in gender relations that are grounded in an ahistorical, mechanistic and static biologically determinist paradigm. As important is the need to dislodge the traditional integration of women as a category, under the general rubric of a universal humanity. Engendering the central paradigms of development and disaster preparedness and management, necessitates a recognition and incorporation into theory and practice, of the following that:
52. Quite obviously, the gender approach though woman-focussed, is not woman-exclusive. It addresses differential male and female roles, activities, conduct, needs, and interests in an attempt to rectify imbalance weighted against women, eliminate hierarchies and usher in mutuality. For to be woman exclusive would isolate women from mainstream development processes, leave male consciousness and practices not addressed and undermine the move towards empowering relationships of partnership.
53. How do we realize this? One means is by mainstreaming, that involves two separate but interlinked processes - mainstreaming women and mainstreaming gender. Mainstreaming women is essentially `political' and ensures their representation and active participation, particularly in policy and decision-making bodies and fora. Women's participation is essential because due to their different gender roles, women have different needs and priorities from those of men.
54. But to achieve greater participation by women, the environment within which activities take place and decisions are made may need to be adapted and made gender-responsive, by mainstreaming gender. This involves:
55. In recent years, new legislation in several countries has established organizations and arrangements for emergency response, specifying roles and responsibilities of various agencies in pre and post emergency situations. Some countries have administrative orders specifying availability of funds at different administrative levels and directives regarding their use. While this paper has not explored these in any depth so as to offer substantive comment, it is necessary for gender progressive groups to catalyse the incorporation of gender sensitive perspectives into these legal formulations and administrative directives; the building into these formulations adequate mechanisms for accountability of implementing bodies and individuals; the provision of legal literacy to women on the existence of these systems, and how women can avail of them and ensure efficient enforcement.
56. Gendered constructions of masculinity and femininity need to change. It is necessary to develop more androgenous personalities, a greater reciprocity in gender relations in every aspect of life, sharing of responsibilities and a greater diffusion in roles. This has been initiated in small measure in the domain of disasters, where creative gender interventions by NGOs are slowly providing momentum to the transformation process. While it would be a travesty of truth to say that creative practice by the CPP and the BDRCS (both engaged in disaster preparedness and management work), have radically restructured gender relations, visible changes in this direction are obvious. That female CPP volunteers in Cox's Bazar Town, disseminated warning information in the 1997 cyclone, and helped women and young children to reach the cyclone shelter, when the evacuation signal was flashed, certainly goes against traditional grain. This is a transgression of conventional domestic boundaries for women, and also breaks the traditional gendered division of labor in disaster work, marking the entry of women into the dangerous male space of disasters. Or for that matter, in the 1997 cyclone (unlike in 1991) many men are reported to have helped women in what is conventionally deemed `women's work' - gathering belongings, herding cattle carrying children to cyclone shelters. Village women and BDRCS field volunteers report that this is largely the outcome of community awareness raised through village level meetings, discussions and gender sensitivity training on the needs and concerns of women and men in disaster and the significance of mutual help, regardless of gender roles.
57. Practices such as these with transformation potential need to be discussed within communities, become more widespread, encompass other areas of life and get institutionalized in the interests of sustained gender justice and partnership.
58. Purely technocratic paradigms of disaster management; an excessive emphasis on post disaster relief operations; a top down gender blind approach that treats affected communities as victims and passive beneficiaries, fails to address the felt needs of vulnerable communities (including women), ignores local resources and capacities, may result in inequitable and unsustainable results, increases vulnerabilities and thwarts disaster resilient development.
59. It is imperative to introduce institutional shifts towards vulnerability and capacity analysis; a concern with disaster mitigation and reduction; a community-based approach that responds to affected populations as `partners in development'; and gender planning and analysis that addresses the vulnerabilities and capacities of women in relation to men in disasters, with a view to innovative gender interventions.
60. Critically linked to women's capacity enhancement in disaster and in its aftermath is their access to support services and infrastructure facilities. As noted earlier, there are significant gender (in addition to class) disabilities and inequities associated with access to shelters, conditions in shelters for women, access to relief items and reconstruction inputs such as access to land, houses or housing material, credit, production or income generation inputs and services, technical information and training on housing reconstruction and income generation and production technologies. Poor women, especially poor female-headed households are particularly hard hit on all these counts.
61. There is clearly a need for a systematic effort to eliminate prevailing biases in the delivery mechanisms of government infrastructure. A greater female presence in disaster preparedness, management, mitigation, relief, rehabilitation and reconstruction activity would certainly be of help in reducing some of these gender biases, but it is as important to reorient these systems, so that even male functionaries recognize the significance of women's needs and concerns and the importance of assisting women. Further while dependence on the state alone is insufficient and more limited in its potential success for reaching women, in comparison to non-governmental organizations, the latter also need a significant re-orientation in regard to incorporating a gender sensitive perspective into their delivery mechanisms.
62. The fore-mentioned legal, social, institutional and infrastructure changes require to be initiated and sustained by the committed engagement of a combination of actors. These include the government, political parties, NGOs and most importantly the community of local women.
63. Collective action by women can radically transform social attitudes and practices that deny women their legitimate material and non-material rights and entitlements.
64. Building group support among and for women, both locally and nationally, is critical for women's empowerment. Such support can come from external groups that provide specialized help to village women; from organizations of rural women themselves and from reorienting men in government and in the bureaucracy, in disaster-related NGOs and men in the local community, towards women's issues and concerns. The presence of women with a gender perspective in decision making roles and positions of authority, both locally and nationally, also has a wider impact on women especially in purdah-practising communities where women are more likely to take their grievances to women representatives than to all-male bodies.
65. In conclusion it may be said that disasters provide an opportunity for disaster preparedness and management theory and practice to actively foreground women's vulnerabilities and to concretely and creatively address these. It also provides the opportunity to create visibility and build on women's strengths and capacities and to institutionalize creative non-traditional gender interventions. The engendering of the central paradigms of development and disasters would contribute to gender justice in disaster response, enhance women's capacities as survivors and ensure their long-term empowerment, thus enhancing sustainable development of their communities.
1. Agarwal, B. (1998) "Disinherited Peasants, Disadvantaged Workers: A Gender Perspective on Land and Livelihood". Economic and Political Weekly, March 28, A-2-14.
2. Bari, Farzana. 1996 "Gender and Disaster: A Case Study, Pattan: Working with Riverine Communities." Duryog Nivaran Proceedings of Workshop on Gender and Disasters, Multan Pakistan, 6-8 March.
3. Cannon, T. 1994. "Vulnerability Analysis and the Explanation of `Natural' Disasters" in Disasters, Development and Environment edited by Ann Varley. Chichester. John Viley and Sons.
4. Corner, L. 1999. "A Gender Approach to the Advancement of Women: Handout and Notes for Gender Workshops, UNIFEM East and Southeast Asia, Bangkok.
5. Cuny, Fredrick C. 1983. Disasters and Development. New York: Oxford University Press.
6. D'Cunha, J. 1997. "Engendering Disaster Preparedness and Management". Lead Article in Asian Disaster Management News: A Newsletter of the Disaster Management Community in Asia and the Pacific, 3(3) November.
7. Enarson, Elaine. 1997. "From the Margins to the Center: Women, Global Disaster, and Sustainable Development". Frap-Net News: A Publication of the Feminist Researchers for Action and Policy Network 3(11) April.
8. Guarnizo C.C. 1992. "Living with Hazards: Communities' Adjustment Mechanisms in Developing Countries" Pp. 93-106 in Environmental Management and Urban Vulnerability. World Bank Discussion Paper No. 168, World Bank, Washington D.C. edited by A. Kreimer and M. Munasinghe.
9. Haidar, Rana. Rahman, Atiq, A. Haq, Saleemul. (Eds) 1991. Cyclone '91: An Environmental and Perceptional Study. Dhaka, Bangladesh: Bangladesh Centre for Advanced Studies.
10. Hewitt, K. 1983b. "The Idea of Calamity in a Technocratic Age". Pp. 3-32 in Interpretations of Calamity from the Viewpoint of Human Ecology edited by K. Hewitt Boston.: Allen and Unwin.
11. Mahbub ul Haq. 1997. Human Development in South Asia, Karachi: Oxford University Press.
12. Maskrey Andrew. 1989. Disaster Mitigation: A Community-based Approach Oxford: Oxfam.
13. O' Keefe, P., Westgate, K. and Wisner, B. 1976. "Taking the Naturalness Out of Disasters." Nature 260:566-7.
14. Susman, P. O'Keefe, P. and Wisner, B. 1983. "Global Disasters, A Radical Interpretation". Pp. 263-83. in Interpretations of Calamity from the Viewpoint of Human Ecology edited by K. Hewitt. Boston.: Allen and Unwin.
15. Torry, W.I. 1986. "Economic Development, Drought and Famines: Some Limitations of Dependency Explanations". GeoJournal 12 (1):5-18.
16. UNDP. Bangladesh. 1999. National Gender Profile.
* Prepared by Jean D'Cunha, UNIFEM, East and Southeast Asia Regional Office, Bangkok. The views expressed in this paper are those of the author's and not necessarily those of FAO.
1 A brief note on the nature of, timing and conditions in which the cyclones of 1991 and 1997 occurred, would further contextualize their gendered impacts, The cyclone of the 29th April 1991 has been described as a super cyclone, whose magnitude was twice the size of the country; it occurred at night and was accompanied by a surge. The last cyclonic disaster of catastrophic proportions in Bangladesh had occurred in 1970, and was hazy in collective memory. Moreover the level of community based cyclone preparedness and management was poorly developed in 1991. By contrast the cyclone of May 1997 occurred during the day, had no accompanying surge and occurred in a context in which cyclone preparedness and management measures, including measures with a gender responsive orientation were much better developed by the Cyclone Preparedness Programme (CPP) and NGOs engaged in disaster preparedness and management in some areas. Also with the horrific memories of the cyclone of 1991 still fresh in the collective consciousness, the affected communities were far more willing to take preparedness measures in 1997. All these factors configured to contain the number of deaths and casualities of the 1997 cyclonic disaster in all the selected areas and reduce the acuteness of its gendered impacts. Despite this however, disaster workers encountered women unwilling to move without their men-folk and families who stayed back to guard their property.
2 The was the result of a study undertaken by this author in phases between August 1997 and October 1999 in six populous and disaster-prone coastal villages of three sub-districts in Cox's Bazar district in South Bangladesh: Chandaliyapara and Katabonia in the sub-district of Teknaf; Charpara and Fakirakata in the sub-district of Moiscal; and Badarkhali and Napithkhali in the sub-district of Chakaria. See Mainstreaming Gender in Disaster Management: The Case of Bangladesh, UNIFEM, (Monograph) fc.
3 I am grateful to the International Federation of the Red Cross Society (IFRCS) & the Bangladesh Red rescent Society (BDRCS), in particular Dr. Purnima Chattopadhayay-Dutt, German Red Cross Delegate, Bangladesh for facilitatating this field research in BDRCS project areas; to the management/staff of IFRCS, BDRCS, the Government Cyclone Preparedness Programme , NGOs & the rural communities for their valuable insights; to Gender for a small grant for a preliminary field scan and to Dr Lorraine Corner, UNIFEM and Loy Rego ADPC for their incisive comments on the paper.
4 More than half of the poor suffer from malnutrition. Infant mortality is 91 per 1,000 live births. Almost 2/3rds of the adult population is illiterate. About 5 per cent of the population have no access to health services. See Mahbub ul Haq. 1997. Human Development in South Asia, Karachi: Oxford University Press.
5 Social factors such as class, religion, rural-urban location, region, etc. breed variations in the condition and position of different segments of Bangladeshi women in relation to one another and in relation to men in different social locations. But macro indicators do suggest women's general marginalization. For details of these see National Gender Profile, UNDP, Bangladesh, 1999.
In the six villages, while education levels were generally low for both men and women, they were significantly lower for women. In Katabonia, of the 35 women interviewed, 30 (88 per cent) were absolutely illiterate, four had primary education, and only one had secondary education. By contrast, of the 27 men in the same area, 14 had secondary education, and 5 primary education, and only eight (30 per cent) were illiterate. In Charapara of 13 women interviewed, 5 had secondary education, and the rest were illiterate. By contrast of the 17 men interviewed, 8 had secondary education, 7 primary education and only 2 were illiterate. It must however be mentioned that the levels of openness and participation for women are higher in Moiscal and Chakaria, than in Teknaf because these areas are better mainstreamed and have a concentration of NGO activity.
6 This is a typical work profile for poor women without assets. Women from land-owning households engage poorer women for post-harvest agricultural operations, poultry/livestock raising and other forms of domestic labour. However though women's class position is seen as derived from her husband's, though women from elite households may enjoy higher living standards or may exercise power/control over poorer men and women, they are in a precarious position. Their lack of independent ownership over, access to and control over resources could render them impoverished and destitute in the event of widowhood, either because they are disinherited by husbands, and/or their share in marital property is appropriated by relatives or if divorced or deserted are denied support by the husband.
7 Women headed households are increasing, especially among the poorest. This is caused by death; male out migration for jobs; erosion of the extended family system due to heightening impoverishment and the consequent undermining of the basis of family solidarity and support; the pervasiveness of the cash economy and the emerging dowry system that causes particularly poorer men to marry more than once, so as to acquire wealth, resulting in increases in the rates of divorce and desertion faced by poor and landless women now than before. See National Gender Profile, UNDP, Bangladesh, 1999.
8 Practical gender needs refer to women's needs in terms of their current gender roles, leaving the status quo in gender relations unchallenged. Strategic gender interests acknowledge the longer term need to change gender roles so that women and men share more equally both the responsibility for domestic and reproductive work and ownership, control over and access to resources and benefits. Strategic interests can only be defined in a comparative perspective in relation to men. Strategies to meet women's strategic interests are designed to raise the status of women relative to that of men.
9 For this classification, see Agarwal, B. (1998) Disinherited Peasants, Disadvantaged Workers: A Gender Perspective on Land and Livelihood, Economic and Political Weekly, March 28, A-2-14.