Women’s Empowerment as a Protective Factor Against Intimate Partner Violence in Bangladesh: A Qualitative Exploration of the Process and Limitations of Its Influence

Abstract

Literature on the relationship between intimate partner violence (IPV) against women and women’s empowerment is contradictory. Findings from a recent survey in rural Bangladesh suggest that empowerment is becoming protective even though IPV rates remain high. We construct qualitative case studies exploring factors and social processes underlying relationships between empowerment and IPV in four villages. Empowerment may be protective against IPV in the aggregate, but this relationship can be subverted at the micro level. Interventions are needed to reinforce the potential of empowerment to reduce IPV and counteract factors such as geographic isolation and limited employment opportunities that inhibit empowerment.

Keywords

empowerment of women gender intimate partner violence

Introduction

Worldwide, in 2010, an estimated 30% of women ages 15 and above had experienced physical or sexual partner violence or both in their lifetimes, based on an analysis of data from 141 studies in 81 countries (Devries, Mak, & Garcia-Moreno, 2013). Intimate partner violence (IPV) is associated with a wide range of adverse consequences for women, including physical and psychological trauma, disability, neurological injuries, and psychological health issues. It also has been associated with gynecological and reproductive health problems including sexually transmitted infections as well as chronic pain and stress (Campbell et al., 2002). Some of the highest rates of lifetime IPV have been reported in rural Bangladesh, where 32-72% of married women below age 50 have reported experiencing it in their lifetimes, and 16-54% within the prior year (Mannan, 2004; Bates, Schuler, Islam, & Islam, 2004; Kishor & Johnson, 2004; Naved & Persson, 2005). In an analysis of the Demographic and Health Surveys (DHS) conducted between 2005 and 2012, Bangladesh ranked second only to the Republic of Congo in IPV prevalence (Solotaroff & Pande, 2014).

Although expressed in various ways, most definitions of women’s empowerment involve their acquisition of resources, agency, and the ability to make strategic life choices in the context of gender inequality (Malhotra & Schuler, 2005). Theories and research findings regarding the relationship between a woman’s empowerment and her risk of experiencing IPV appear contradictory. Some researchers have found that an unempowered woman’s economic and social dependence on her male partner puts her at particular risk of experiencing IPV because she perceives and has few alternatives to the abusive relationship (Dobash & Dobash, 1979; Kalmuss & Strauss, 1982). Others suggest that empowerment may exacerbate a woman’s risk of IPV, because she is seen as challenging gender norms and as threatening her male partner’s dominant status or exercise of power (Atkinson, Greenstein, & Lang, 2005; Choi & Ting, 2008; Counts, Brown, & Campbell, 1992; Jewkes, Levin, & Penn-Kekana, 2002; Macmillan & Gartner, 1999; Schuler, Hashemi, & Badal, 1998). The nature of the relationship is likely to be time and place specific, contingent on the normativity of women’s empowerment and the extent to which it can be leveraged to change women’s options, both inside and outside marriage (Ackerson, Kawachi, Barbeau, & Subramanian, 2008; Koenig, Ahmed, Hossain, & Khorshed Alam Mozumder, 2003; Naved & Persson, 2005).

In a review of published data from 41 settings (Vyas & Watts, 2009), household assets and women’s education were generally protective against IPV, but evidence about women’s involvement in income generation was mixed. Similarly, in other studies, a woman’s control over resources appears to be protective (Dalal, 2011), but participation in household decision making (Rahman, Hoque, & Makinoda, 2011) or earning an income and contributing more than a nominal amount of income to household expenses have been shown to put a woman at greater risk of IPV (Anderson, 1997; Hadi, 2000). A recent, comprehensive review of published evidence on causes of IPV (Heise, 2011) found that women’s education was protective but that other factors related to women’s empowerment were protective in some settings and deleterious in others, including women’s employment, women’s participation in credit or other development programs, and women’s asset ownership. The same review also identified a number of gender norms that were associated with increased risk of IPV, such as a cultural emphasis on women’s purity and family honor, acceptance of wife beating and a man’s right to discipline/control female behavior, and gender hierarchical or transitional attitudes among men. The review also identified IPV risk factors related to spousal conflicts over gender norms, including “patriarchal triggers” such as challenges to male authority, assertions of female autonomy, and the failure of either spouse to meet gender role expectations.

In Bangladesh, where empowerment of women appears to be increasing but IPV prevalence remains high, the association between empowerment and IPV appears similarly complicated. Probable catalysts in the trend toward greater women’s empowerment include rapid increases in women’s and girls’ schooling levels (Malhotra & Schuler, 2005; Schuler, 2007; World Bank, 2003), which can lead to dramatic changes in women’s roles and aspirations (Schuler, 2007), the garment industry (Amin, Diamond, Naved, & Newby, 1998; Kabeer, 2001), and women’s increased engagement in agriculture in conjunction with its intensification and diversification (Jaim & Mahabub, 2011). Nongovernmental organizations (NGOs) that create economic and social opportunities, drawing women by providing microcredit, training, employment, and legal education and services are also believed to be contributing to empowerment of women (Jahan, 2012; World Bank, 2006). An analysis of DHS data from 2007 identified two specific empowerment of women-related factors associated with an increased risk of IPV among women in Bangladesh: women’s income-earning work and their involvement in nonfamily institutions (Amin, Khan, Rahman, & Naved, 2013). Other studies in rural Bangladesh suggest that some forms of empowerment may diminish a woman’s risk of IPV by relieving economic pressures on the household, enhancing the woman’s perceived value in her family, and increasing her economic self-reliance, public visibility, access to social support, knowledge, and skills, and therefore her options for leaving an abusive relationship (Hashemi, Schuler, & Riley, 1996; Schuler & Hashemi, 1994; Schuler, Hashemi, Riley, & Akhter, 1996; Schuler, Nazneen, & Lemons, Unpublished results).

In Amin and colleagues’ (2013) analysis, women’s education, both at the individual and at the community level was found to be protective, but other studies in Bangladesh have shown that the relationship between individual- and community-level women’s empowerment and IPV is highly context specific. Koenig and colleagues (2003) found that higher individual-level women’s autonomy and short-term membership in credit groups were both associated with elevated risk of IPV in more culturally conservative areas, whereas in a less culturally conservative area, individual-level women’s status indicators were unrelated to the risk of violence, and community-level measures of women’s status were associated with significantly lower risk of IPV. Similarly, Naved and Persson (2005) found that whereas some factors were significantly associated with increased or decreased risk of violence in both rural and urban women, others were only significantly associated in one site, including income earning, education, and participation in savings and credit groups.

Findings from a recent, nationally representative sample of married women suggest that women’s empowerment is becoming a protective factor against IPV in this setting even though rates of IPV remain high (Bates, Osypuk, Naved, Yount, & Schuler, 2015). Empowerment of women was found to be protective both at the individual and the village levels, but it did not hold in every community included in the study sample. Building on conceptual frameworks posed by Heise (1998), Koenig and colleagues (2003), and Naved and Persson (2005), this article uses qualitative data to examine some of the ways in which empowerment of women can interact with various contextual factors to discourage IPV in some settings and support its persistence in others. These models posit that IPV is a function of the interrelated effects of contextual and community-level factors, household and individual-level characteristics, and women’s status/autonomy factors; the authors further theorize that an individual woman’s empowerment changes from a risk factor to a protective factor against IPV as gender norms become more egalitarian, but that this change occurs more or less rapidly depending on contextual factors that vary at the local level.

Method

Site Selection

We collected qualitative data from four communities selected purposively from a nationally representative sample of 78 communities where surveys were conducted in 2013 and 2014 as part of a larger, mixed-method, multilevel study on relationships between empowerment of women and IPV. To maximize variation in the relationships of interest, we used the 2013 survey data to identify four communities with high or low levels of IPV and empowerment of women by categorizing them into four cells and choosing a village from each cell that was far from the mean on both variables (Table 1). Our index of empowerment of women for categorizing the villages included eight dichotomous indicators of empowerment described in a previous study of rural credit programs in Bangladesh (Hashemi et al., 1996), each tapping a potentially separate dimension of empowerment (e.g., mobility, involvement in family decisions, economic security).

Table 1.

Village-Level Mean Past-Year Physical IPV Prevalence and Empowerment Scores Among Married Women in Sampled Villages, 2013.

Past year IPV %	Empowerment score
Past year IPV %	High	Low
High	Village B: 51%/.28	Village A: 65%/−.14
Low	Village C: 15%/.42	Village D: 21%/−.43

Note. IPV = intimate partner violence.

Data

We conducted a total of 16 focus group discussions (FGDs)—seven with men (36 participants, aged 27-62), nine with women (45 participants, aged 16-65)—10 in-depth interviews (IDIs) with men (aged 35-65), and 16 life history narrative (LHN) interviews with women (aged 18-30), with approximately equal numbers from each community.

We developed interview guides for each type of interview to ensure that key topics would be covered, but the interviewers were encouraged to be spontaneous and add follow-up questions based on each study participant’s remarks. The women had to be at least somewhat empowered based on their incomes, education, role in managing family resources, lower-middle or lower class (based on the interviewers’ assessment), and married fewer than 12 years (a criterion intended to best capture experiences of women thought to be at highest risk for IPV). They were identified by means of key informants and informal screening interviews. The LHNs focused for the most part on the women’s own experiences, whereas the FGDs and IDIs primarily addressed trends in the communities. The interviewers were two Bangladeshi women and one man with long experience in conducting qualitative interviews. They received specific training for this study, and they contributed to the design of the interview guides. In each village, the interview team conducted an informal participatory mapping exercise with community members to identify factors that may influence empowerment of women and IPV, including infrastructural developments, the presence and history of NGOs in the village, and landmarks such as waterways, agricultural areas, and neighborhoods. The interviewers compiled field notes based on observations and casual conversations. The interviews were conducted in Bangla, transcribed, and translated into English.

Analysis

In the initial analysis, the first author read and reread the transcripts as individual and group narratives, aggregating them by village to identify themes and evidence regarding the relationships between empowerment of women and IPV. She then discussed the themes with the field research team in light of their village maps and field notes and invited them to offer explanations for the commonalities and differences observed among the villages with respect to empowerment of women, IPV, and relationships between the two. We subsequently used NVivo 10 to code the transcripts thematically, and systematically reviewed and summarized the evidence and counter-evidence pertaining to each theme. The study was reviewed and approved by the Institutional Review Boards of FHI 360 and BRAC University.

Limitations

As with all qualitative studies, the results of this study cannot be assumed to represent any larger population beyond the four villages from which the data were collected. However, we collected enough information about the villages to establish that none of them is particularly unusual in the rural Bangladesh context. A second limitation is that the data represent a snapshot in time; as we do not have similar interviews from an earlier time period, we must assume that observations about economic and cultural changes over time have some basis in reality. However, we recognize that there is often a tendency to idealize the past and to exaggerate descriptions of change. We considered this in our analysis.

Findings

Some of the same themes emerged in all four villages. People spoke about changes underway in the country at large, or the larger geographic area in which their villages were located, changes they had seen firsthand or learned about from labor migrants, television, NGOs, and others visiting the villages. We did not try to prevent people from discussing changes they observed elsewhere that may have been infrequently seen in their own villages, as these were of considerable interest to them, but we did repeatedly ask them to tell us specifically about their own villages. Despite differences in relative levels of empowerment of women and IPV, all four villages shared some characteristics with each other and rural Bangladesh, in general, such as being predominantly Sunni Muslim, continued predominance of arranged marriages and dowry exchange, and largely agrarian livelihoods. In the village-by-village descriptions that follow, we have tried to identify and focus on themes that were site specific in terms of relative importance and greater emphasis.

High Empowerment of Women, Low IPV (Village C)

Village C, only 2-2.5 hr distance from the capital, Dhaka, by road, consists mainly of people who were resettled after devastating floods in 1988 and 1998. Thirty-two households belonged to wealthy families who owned large parcels of land and had resettled in Dhaka or abroad, having leased their land to poorer families or given it out to sharecroppers. Many of their homes were occupied by the tenants. Patron–client relationships were strong and cordial. Several of the wealthier families who had relocated in Dhaka ran garment factories or construction businesses in which they employed young people from the village. On religious holidays, they distributed clothing, food, and cash in the village and provided support for medical treatments. Frequent contact with these elite families, as well as with poorer labor migrants who went back and forth to Dhaka, had fostered a desire to mimic the perceived norms of the educated middle class. Many parents said they hoped that, with education, their children could become bhadrolok, or gentle, cultivated people who spoke and behaved nicely.

Besides farming, men from poorer households were pursuing a mix of occupations, including bicycle-rickshaw and “van” pulling, carpentry, and construction of brick houses. Nearly 100 individuals, mostly men, did low-paid jobs in Dhaka, returning to the village once or twice a month. Women were engaged in a variety of economic activities outside the home, including excavation of earth for building and road construction, and weeding, transplanting, harvesting, and post-harvest work in the cultivation of various crops. Most of them handed over their earnings to their husbands. Some 15-20 unmarried girls were working in garment factories in Dhaka, sending money back to their parents.

Several NGOs were active in the village in health, education, and microcredit. Village C stands out in having several young married women with university education who were admired and seen as role models. One such woman was teaching in a school run by BRAC, a large NGO. Her husband was encouraging her to go back to school to get her master’s degree. Another had completed 2 years of university. Her husband too was encouraging her to study further so she could get a formal job. Both of these women were from poor households. In one FGD, a group of illiterate women spoke about the value of education. One of them hoped her daughter could get educated and become a doctor:

At the hospital the girls that are there, the doctors (some may have been nurses or paramedics) seem like they are barely 15-18 years old! All are the same size—so cute. . . . Seeing that, it resonates in my mind that I would be at peace if my daughter could educate herself and become like them! (36-year-old woman, started primary school)

Some women demonstrated their empowerment by financing and making decisions about their husbands’ work.

I brought in 30,000 taka from my parent’s house for my husband . . . to start a new business . . . my husband was an employee in a shop. . . . I told my husband I wanted to talk directly to the owner of the shop. . . . I asked him directly: “Brother, are you really willing to sell the shop? Why do you want to do this?” (26-year-old woman, started secondary school)

She bought the shop, then bought a cow, and “flipped” it a few days later, when the price went up just before the Eid festival, and gave the money to her husband to expand the business. But her husband decided to go abroad to work and the shop languished while he worked out the details. After he failed to get a visa, she helped him restart the shop, but first spoke with nearby shop owners to assess his seriousness. When she decided to give him one more chance, she threatened him, saying,

If you again incur a loss I will get the money back by selling off your kidney.

This is an unusual woman. Many educated and/or economically active women in this village took care to show at least symbolic adherence to traditional gender norms, and both men and women mostly rejected the suggestion that women were becoming more powerful than men. As one woman put it (she had completed 2 years of university education whereas her husband had only a seventh grade education), she was careful not to “dwarf” her husband. In other ways, however, the case above illustrates several gender norm-related behaviors that had begun to change in this village: Women felt comfortable in the marketplace, they had become articulate and confident enough to converse with men who were not kin, they had access to resources that they used in negotiating with their husbands, and they followed their own judgment when they believed they were right.

Both men and women in Village C said that, because of their earnings, women now had the option to leave abusive marriages, or they could refuse to work. At least one abused wife had left her husband with their four children. Men spoke of mutuality and the need for cooperation between spouses to thrive economically. There were several examples of men doing housework when their wives were busy (although most of the burden still fell upon women). These men did not seem concerned about adhering to the traditional stereotype of the man as breadwinner and the woman as homemaker. A woman in a FGD observed,

In the past, men would work in the fields. They wanted to have [their food] ready when they returned from the field. . . . In case the woman failed to get things ready, he would beat her. . . . Now the situation has changed. A husband today would not start beating his wife if he did not see everything ready when he came back from the field. Rather, he would try to help her in preparing things. (35-year-old woman, started secondary school)

A woman in the same FGD summarized the change in gender norms in dramatic terms:

In the past women were like dolls. They did not have any power or value. But times have changed, it’s a new age. Now women have rights and access to everything in the family. They have the same value as their husbands. (32-year-old woman, completed primary school)

Despite these changes, a few, mostly older, men still considered it shameful for women to work outside the home:

Women coming home from the fields with a sack full of things are not good to see. I cannot like it. This is indecent to me. I cannot think of my wife carrying such a load while I am alive. (55-year-old man, illiterate)

Many women gave examples of strategies they had adopted to conform to traditional norms or to minimize the appearance of going against the norms, including choosing income-generating activities that were home based; handing over their earnings to their husbands or fathers-in-law; scrupulously accounting for every taka they spent; voluntarily discontinuing their education, a job, or visits to friends because their husbands did not approve; or maintaining that what they did was just what everyone else did or was for the benefit of the children.

And, surprisingly, even the most empowered women in Village C still condoned IPV to some extent, even if they themselves did not tolerate it. Or they pretended to condone it in situations where the woman was described as being “at fault” based on traditional gender norms. We are unable to judge whether they really meant that they believed IPV to be justified in such situations, or whether this was an example of paying lip service to traditional gender norms that they did not follow themselves.

In summary, in Village C, women’s increasing empowerment was widely recognized. Men saw its benefits and believed wives could now leave their marriages if mistreated. People were also exposed to nonviolent values they attributed to the educated, urban middle class, which they associated with success and benevolence. However, many women continued to pay lip service to traditional gender norms.

Low Empowerment of Women, High IPV (Village A)

Similar to Village C, Village A, one of the two poorest of the four, consists mainly of people who were resettled after losing their homes and fields to river erosion. Households are separated into three geographically separate clusters strung along a metaled road that was constructed to protect the area from monsoon flooding. The poorest live on the river side of the road. One cluster of houses, created as part of a governmental rehabilitation project was built on tiny plots of land with no space for women to engage in home-based economic activities such as poultry and livestock rearing or fruit and vegetable cultivation. Lacking other opportunities, some of the women from these households were employed by the World Food Program and CARE in road maintenance work. Many of the houses in a second cluster were locked up, the inhabitants having immigrated to Dhaka for employment. Men in the remaining households earned their living through a mix of agriculture, daily wage work, and temporary employment in Dhaka. Quite a few women as well were working in Dhaka as maids or in garment factories. As the soil tends to be sandy from flooding, in most fields it was possible to plant only one crop per year, and agriculture provided few opportunities for women’s employment.

Lacking nearby transportation links, the village is relatively isolated and women’s mobility was low. Because of poverty, only a few families owned television sets. Men watched TV in tea stalls, but most women lacked access. In addition, social cohesion was low because the village clusters were isolated from one another. These factors created obstacles for women to build social support mechanisms, and to become exposed to alternative social norms and acquire income earning and other skills through contact with others.

Various NGOs worked in Village A. One NGO had been engaged in consciousness raising and political organizing related to class and gender exploitation for approximately 25 years but these efforts had lost their intensity. Women had some legal knowledge and were able to speak articulately about issues such as women’s rights, but in most cases, their incomes were meager and they had minimal decision-making power within their families. Despite the NGO’s attempts to instill ideas about human rights, some of the interviews with men revealed considerable resentment of women who were seen as violating gender norms, with gender norms sometimes couched in religious ideology. The man quoted below (53 years, illiterate) admitted that women were helping their families to prosper, but he said they had become aggressive (chora) and no longer respected and honored men as they used to.

These changes [in women’s mobility and work] have not been positive for the society, because women should not move about in an unruly manner. . . . We live our life according to Sharia law. If we are to abide by that law, it is wrong for women to frequent the markets and other places.

In the past, if a woman saw a man standing somewhere and he was casting a shadow, she wouldn’t even cross that shadow, she would take an alternate route to her destination. Now a woman wouldn’t mind bumping into you as you pass. If she needs to, she may well shove you out of her way and go ahead of you.

These themes were also echoed in a number of FGDs with men:

Q: As the power of women has increased, have you men ever felt that your own power is not as it was before, that it has decreased?

Participant 4: Many of us may have thought this, but one cannot express it. One burns to ash inside but cannot express it . . . (47-year-old man, started secondary school)

Q: You friends hang out here at the tea stall and talk—you can’t talk about this?

Participant 2: Many discuss it . . . (42-year-old man, illiterate)

There was some evidence that women’s ideational empowerment with only moderate economic empowerment to back it up had become a trigger for IPV, as the women sometimes challenged their husbands’ authority but lacked bargaining power.

Q: Did women in the past argue with their husbands, or do today’s women do it more?

Woman 4: Women could not say a word in the past. . . . If husbands say something wrong, today’s women show [them] their mistakes by arguing with them. Sometimes this arguing turns into fighting. (27-year-old woman, completed secondary school—passed SSC exam)

Status disparities were also mentioned as a source of contention between husbands and wives:

The husband has to face too many problems if his wife is educated. He thinks his wife is more educated than he and if she gets a job she will not obey him. There are often arguments between husband and wife for this reason. (26-year-old woman, started secondary school)

Women in an FGD said that they, or their society, condoned beating of women who argued with their husbands or failed to be submissive:

Woman 4: It is okay to beat wives who become very undisciplined. (27-year-old woman, completed secondary school—passed SSC exam)

Woman 2: Being unsubmissive to their husbands, if they go here and there, then if the husbands beat them society will say that it is okay. They have become disobedient to their husbands. (55-year-old woman, illiterate)

Woman 1: Often wives argue a lot with their husbands. Then if the husbands beat their wives people say it is okay. (50-year-old woman, illiterate)

All four of the empowered women from Village A from whom we elicited LHNs said that they had been hit or beaten by their husbands, in two cases with sticks and other objects in the last year. Precipitating factors included (a) failure to get money from her parents to support his gambling, suggesting they buy a cow, urging him to allow her to go to Dhaka to work, and suggesting that he go out to work regularly (he was a daily wage laborer); (b) hearing from neighbors that she had gone to various places or chatted and laughed with other women; (c) arguing with her husband, allowing the baby to cry, and failing to do her housework according to his standards; and (d) having a miscarriage, failing to bring his meal to the field, and going out without his permission. Two LHN participants said they believed they had been at fault every time they were beaten; the other two said their beatings were justified on at least one occasion.

In the two FGDs with women and all the LHNs and IDIs, study participants claimed that IPV had diminished and, indeed, it may have been even higher in the past. Some of the men said it had diminished, but others disagreed:

Man 4: There is more torture [than before]. The younger brother of X beat his wife until she was half dead. She couldn’t even walk to her house. The case went to the Chairman (local elected official). Then the wife went to her father’s house, then her father and brothers brought her back. Women are still being tortured . . . (47-year-old man, started secondary school)

A shalish (traditional mediation group) member said severe IPV was declining, but noted that it was not “in vogue” for a man to be required to beg forgiveness from his wife, even when he was deemed to be at fault for the beating. In contrast, if the shalish found the woman to be at fault, she was supposed to hold her husband’s feet and seek forgiveness.

In summary, in Village A, women’s opportunities for economic empowerment were limited both by lack of homestead land and by poor quality agricultural land in the village. Women were socially empowered enough to challenge traditional gender norms and men felt threatened by this. IPV was common.

Low Empowerment of Women, Low IPV (Village D)

Village D is even more isolated and lacking in social cohesion than Village A. It consists of three clusters (para) some distance apart, linked only by a muddy path that meanders through the rice paddies, which are submerged in the monsoon season. Interactions among paras were minimal. Each has its own mosque and there is no common gathering place. Because of annual flooding, only two crops are planted per year. As in Village A, women’s mobility was limited, in part by the lack of convenient transportation links. Even so, many of the men from Village D and a number of women had found jobs in the district town or in Dhaka, and about 25 people, mostly men, had gone abroad to Middle Eastern countries, Singapore, or the United States. Perhaps because of remittances, Village D was economically better off than the other three. The majority of households were lower-middle class and many people from this village said it was “not necessary” for women to work outside the home. Women who did work outside their homes tended to be from the lowest or the highest economic strata, the latter mainly working in Dhaka or regional towns.

Extended families were more common here than in the other villages, and women whose husbands were absent often lived with their in-laws. Those who were de facto household heads often were in close touch with their husbands via mobile phone, which enabled the men to maintain control of important and, in some cases, even routine decisions. Women who raised cows at home often did not consider themselves to be the owners of the cows and typically did not control the income they earned from this source. They usually did keep the income they earned from raising and selling poultry, but many said poultry were raised mainly for household consumption of eggs. The strength of the traditional gender norm of husband as provider is illustrated by the reported misrepresentations of some wives we interviewed, as reported by another woman who herself had grown up in towns and, therefore, found it strange.

I know that the women living over there all sell eggs, chickens and ducklings, but they did not tell you about that, out of shame. . . . I said to him [her husband], “If I earn some money, what is the problem? . . . Our household is new and we don’t have anything. . . . We do not have a sheet on our bed, do not have a single pillow with a cover—where shall I sleep?” In reply he said, “It’s okay, I will buy you those things after I get a job. . . . If women have money, they get onto their husbands’ heads, they do not obey their husbands.” (18-year-old woman, started secondary school)

Once when her husband beat her, her sisters-in-law advised her to be more subservient:

It was your fault. Why do you argue with your husband? We never argue with our husbands and our husbands never beat us.

In traditional mediation over IPV, according to men in an FGD, if the man is found guilty, he is obliged to seek forgiveness from the village elders (not from his wife). If the husband claims he had a good reason to hit his wife, and she, or witnesses, say her husband’s claim is true, she is warned not to make the same mistake again. The husband is not disciplined.

Most women in Village D, like those in Village A, had not achieved a level of economic empowerment that could deter IPV, but their adherence to traditional gender norms seems to have protected many women from IPV, and there is evidence that those who deviated were sometimes beaten for trying to earn an income. Although there was little apparent influence of urban culture on gender norms, people characterized urban culture as civilized and refined, and in that context described IPV as shameful, as in these statements from two FGDs with women and the following with men:

Q: What is the reason behind the decrease in torture?

Woman 3: Well, you know, everyone cares about their honor and dignity. (55-year-old woman, illiterate)

Woman 4: I think education is the main reason why torture of women has decreased. It is also the reason that manners have improved. . . . By going to school, they get to socialize with educated people and thus they learn. (35-year-old woman, started primary school)

People are becoming conscious. It would be a disgraceful situation if everyone in the para knew they were fighting. Because of this, men do not beat their wives. As people become educated, fear of disgrace increases. (55-year-old man, illiterate)

People go to cities. They observe how educated people behave with their wives. Gradually they gain the ability to control themselves from hitting their wives. (50-year-old man, completed higher secondary school—passed IA exam)

As in Village A, people also said that husbands’ frequent absences deterred IPV.

Thus, in Village D, as in Village A, women were not very empowered economically but as poverty was less acute, women were able to embody the traditional gender norms of low mobility, low economic productivity, and dependence on their husbands. Other factors discouraging IPV included man’s absences due to migratory labor and social stigma associated with IPV.

High Empowerment of Women, High IPV (Village B)

Village B is another relatively isolated village in which a large majority of the households are poor. Electricity only became available recently, and only in one section of the village. Although most households did not own land, triple cropping created a substantial demand for agricultural labor both in the village itself and in nearby areas. Vegetable cultivation and milk production had become profitable. A breed of cow that produces larger amounts of milk was introduced by local NGOs and two commercial enterprises, BRAC and Milk Vitae, collected milk daily for export to the subdistrict town. Lack of employment opportunities for men had been a chronic problem, but with prosperity increasing, brick houses were being built, creating further employment for landless men as masons and construction assistants.

Our survey findings indicate a relatively high mean level of empowerment of women but the qualitative research team assessed it as only moderate. Women’s mobility outside the village was limited, but women were able to purchase household goods because there were several shops within the village itself. Many women earned money by raising milk cows and poultry and took their cows out to graze. A handful worked as tailors, or for NGOs. Women were also intensively involved in crop processing after the harvest. Interviews with men suggested that some had come to recognize and appreciate women’s economic contributions, whereas others had not, or were ambivalent:

They (the elders) like to say that in the past no women went outside the house—ever. Now women are joining the society, going here and there. In the past, women never went out of the house but in the past men starved to death too. Now no man is starving to death.

There is a progress of the family, but it looks odd to many men. . . . There might be some harm to the society since if my wife goes outside she will be seen by men. Those who are a bit wise and very religious might say: that man is using his wife for shopping or for taking the cattle to the field and such. (40-year-old man, completed primary school)

They don’t do anything except raising goats, hens, ducks. The women of our area don’t work outside the home. (50-year-old man, completed secondary school—passed SSC exam)

This was a village of competing gender norms. For example, a man with socially and politically liberal views was influential in one section of the village and, until his death 4 years earlier, a traditional religious leader had been influential in another section. One person estimated that half the women kept their heads covered. A young woman who had married into the religious leader’s family had been beaten on various occasions for violating gender norms by attending an NGO training program, raising cows and goats, arguing with her husband, and failing to do her housework to her husband’s satisfaction (traditional gender norms gave men the right to beat their wives for this reason). After describing what she now did to help support her family, she added,

Now the munshi has died and we can leave our houses. . . . That munshi gave us a very hard time! We couldn’t even peep through the door when a hawker came. All men were prohibited for us. Whenever they saw us talking with other men “they” would complain to our husbands and our husbands would hit us. The married women of this neighborhood feared him a lot. . . . In those days if a woman of this neighborhood wanted to buy something from a hawker she would first check to see if the munshi was around. If he wasn’t, she would talk with him and buy things from him inside her house, but if the munshi somehow noticed it she could not escape being beaten by her husband. The munshi would call on that woman’s husband and say: “Your wife has done this! What she did is prohibited (haram)!” (29-year-old woman, started secondary school)

There were many examples of women being beaten for traditional reasons related to prescriptive gender norms, and even some of the more empowered women still accepted this:

When I get beaten I feel ashamed because I got beaten because of my own faults. (25-year-old woman, started primary school)

But there was ample evidence too that gender norms were contested.

Q: So men help their wives with household chores—how do people in the community react?

A: Most people take it positively but some take it negatively (others agree). (22-year-old man, completed secondary school—failed SSC exam)

Q: But don’t elderly people say things like: he has become a “henpecked wussy?” (Interviewer used phrase from another interview)

[Everyone laughs and agrees.]

Q: Who exactly makes such comments?

A: Well, if five people approve of something, there will be another five who disagree. (18-year-old man, started primary school)

Several men said they sometimes helped their wives with domestic chores and that this was natural, because their wives were busy working and earning. Others said such a thing would never happen. Both men and women made many statements about women’s role in household decision making and the desirability of cooperative planning and decision making between husband and wife. Many men and women also stated that women—despite a recent evolution in their economic roles and the fact that some were more educated than their husbands—never dominated their husbands and men always remained in control. Others complained that women had become domineering. They made sarcastic remarks about men who appeared to be dominated by their wives, or even listened to them.

[Sarcastically] It’s the digital age. Husband and wife have equal rights. Government has established that. . . . Of course people say: “You do everything according to your wife’s suggestions—you even take a pee when your wife wants you to!”—they say these bad words. . . . Some men are henpecked (varua). They listen to their wives. This is what they are called in our village. (50-year-old man, completed secondary school—passed SSC exam)

Many of the women in the study, such as those in the following focus group, made statements implying they endorsed traditional gender norms or disapproved of women’s education and their growing mobility and economic roles:

Women had better manners in the past. Today’s women don’t have good manners. (52-year-old man, illiterate)

Some women think, “I am educated. I understand better than they do. My husband is illiterate. My father-in-law and mother-in-law are illiterate. What do they understand? Why should I do things according to them?” Many girls have turned bad after being educated. (48-year-old woman, illiterate)

But the woman quoted directly above later observed,

In the past, women didn’t have any work from which they could earn income, which is why their husbands would hit them whenever they failed to follow their commands properly. Even now a woman who doesn’t earn income by doing some work gets beaten by her husband for the slightest of mistakes.

In summary, this seems to be a village in transition. Women who are not economically empowered are sometimes beaten for traditional reasons, and as they become somewhat empowered they are beaten for challenging gender norms.

The LHN Sample

Although a sample of only 16 empowered women (four per village) is too small to draw conclusions from, it does provide insight into the potential relationship between an individual woman’s level of empowerment and community-level factors such as acceptability of empowerment of women and relative absence/presence of patriarchal norms. For example, in Village C, which was a village with a high level of empowerment of women and low IPV at the community level, only one of the empowered young women had ever experienced minor violence (i.e., slapping), which she attributed to an isolated argument with her husband over his association with known drug users. The women in Village C generally said they were not affected by violence either because their husbands are coolheaded or pious or because they are obedient to their husbands, thus they do not need to be disciplined.

In Village D, which like Village C had generally low levels of IPV but was a low empowerment of women village, three of the four women had been beaten, including one who had experienced several severe beating episodes and had been slapped in the past year. She said her beating was explicitly linked to her empowerment activities and her husband’s belief that she was challenging his status. The women in Village D attributed any decrease in the frequency or severity of IPV to either their husband’s absence or their efforts to appear more obedient, rather than being related to their empowerment characteristics.

He says, “You talk too much, having earned some money by making dresses. If I buy you poultry birds you will talk more after rearing them and feeding me some chicken and eggs. I cannot help you let that happen. . . . ” Later, I bought a hen after going to my father’s house and taking some money from my mother. Because of that he slapped me once for buying the hen without his permission and bringing it home, but he did not beat me more than that. . . . I thought beforehand that he might well get angry with me, so, I did not mind when he slapped me. What is the problem if I can rear two poultry birds after taking a slap? I would benefit in the long run. I bought that hen after careful thought. (18-year-old woman, started secondary school)

Village B was selected because it had a relatively high level of empowerment of women (although lower than Village C), and a high overall rate of IPV; thus, it is not surprising that the rate of IPV was high among the participants (all four had been beaten). However, although in at least some cases their beatings were precipitated by characteristics or activities related to empowerment of women, only one woman said she had been beaten in the last year, whereas the other two to three (one did not specify) were no longer beaten, which would be consistent with the idea that IPV may decrease when a woman reaches a certain level of empowerment. When asked about protective factors, participants’ answers were mixed, with one citing her husband’s newfound appreciation of her contribution to the household as the reason beating had decreased in severity (although she had been slapped in the last year), two citing their obedience to their husbands (including consciously moderating their behavior), and one saying beating had stopped after they entered a separate household and conflicts with her mother-in-law ceased.

I was hit for raising cows and goats. . . . I told him people would not praise us if we starve day after day, they would criticize us. People would not buy us the things we need. They would not sell rice to us at lower price! Yet, he didn’t give his consent. I put up with him cursing and hitting me; but also proved to him that raising goats was profitable and that it earned me enough money to cover family needs. . . . I sold a goat and gave him the money. Then he realized that he should allow me to do these things. Now he buys goats willingly. He also gave me a hard time when I attempted to raise cows. He hit me for going out to buy cow feed. But later, as I managed to buy rice by selling milk, he realized that this was profitable too. . . . He doesn’t hit me now. He hits me less frequently now—but it reduced gradually. (29-year-old woman, started secondary school)

Village A was selected because of its high rate of IPV and its low overall levels of empowerment of women. As expected in a high IPV setting, all four women had experienced violence, including half in the past year. None of these women described their income-generating activities or contributions to household management as protective, and in fact three said they were beaten over incidents related to their empowerment characteristics or activities, including asking for a cow, directing her husband to get a job, and socializing with neighbors. Any decreases in IPV were attributed to their husband’s absence while in Dhaka or their conscious efforts to moderate their behavior (including curbing their mobility and avoiding discussions about money).

Discussion

As shown in Table 2, village case studies suggest several community-level characteristics that may impede or facilitate women’s empowerment, IPV, and the relationship between the two.

Table 2.

Community-Level Factors Related to Empowerment of Women and IPV in Four Villages.

Supporting empowerment of women	Discouraging IPV	Making empowerment of women a risk factor for IPV
•Absence of patriarchal ideology •Access to urban centers •Successful female role models •Women’s economic contributions acknowledged •Women’s education •Women’s employment opportunities	•Aspirations for educated lifestyle •Less poverty •Men’s absence •Mutuality between spouses •Nonviolent male role models •Patriarchal ideology weaker •Women’s adherence/lip service to traditional norms •Women’s economic contributions acknowledged •Women seen as having option to leave husbands	•Men disparaged for adopting “female” roles •Men view empowerment of women as threat to masculinity •Patriarchal ideology •Social without economic empowerment of women •Spousal status disparities favoring women

Note. IPV = intimate partner violence.

Factors appearing to foster women’s empowerment in the study villages include the proximity of the village to Dhaka and district towns; transportation links and markets; economic opportunities for women and demand for women’s labor; women’s education; exposure to socially respectable, prosperous female role models; and the relative absence of patriarchal, religious ideology. Factors appearing to discourage IPV in the four villages include the relative absence of patriarchal ideology; men’s absence (for migratory labor); the absence of severe poverty; the recognition and high valuation of women’s potential and actual economic contributions; women’s lip service to traditional gender norms; the association of collaborative, mutually supportive relationships between spouses with economic success; and the association of nonviolence with education, refinement, and, more generally, a lifestyle to which people aspire.

Although patriarchal gender norms and Islamic religion are the norm in Bangladesh and in all four villages, interviews and field notes indicated that there were differences in the extent to which these norms are embraced and enforced in the four villages. Several decades ago, researchers noted that men often failed to adhere to the patriarchal norm that defined men as providers (Cain, Khanam, & Nahar, 1979); this was still a common source of marital conflict in the1990s, and defensiveness and conflict over this failure often provoked men to abuse their wives (Schuler, Lenzi, Nazneen, & Bates, 2013). Village D is notable because, more than in the other villages, women tend to adhere to traditional gender norms both in their behavior and in the narratives surrounding their behavior. Perhaps because extreme poverty was virtually nonexistent and men were able to actualize the norm of husband as provider, rates of IPV were relatively low despite the low level of empowerment of women.

Overall, the findings (particularly those from Village A) are consistent with the idea that a moderate level of empowerment of women may provoke IPV, as women begin to behave in ways contrary to traditional gender norms, including by earning incomes, moving about in public, building social support networks outside the family, making decisions on their own, and, sometimes, standing up to their husbands, disobeying them, or arguing with them. This is the stage of normative change at which men are sensitive about their own status and vulnerable to social criticism and teasing about their own behavior (as when they help their wives with housework) and their wives’ behavior (as when wives associate with other men). Men’s sensitivity to challenges to their masculinity seems to have persisted in Village B despite a moderately high level of empowerment of women. Women in Village C seemed to understand this risk, and therefore made considerable effort to portray their behavior as consistent with traditional gender norms, and insisted that their husbands were superior to them. The extent to which women’s tendency to condone IPV in their communities (if not in their own relationships) is genuine or merely another example of lip service to patriarchy merits further investigation, as this question has implications for designing behavior change communication and other strategies to address IPV.

The findings suggest that women’s tactics to accommodate gender inequality such as verbally supporting patriarchal gender norms, handing over their incomes to their husbands, trying not to “dwarf” their husbands, and, in some cases, wearing burqas, may provide some protection against IPV, but gender transformative interventions are needed to address the underlying norms that continue to support IPV even as gender roles evolve. Such interventions could accelerate and reinforce the influence of women’s empowerment in reducing IPV, and counteract the effects of factors such as geographic isolation and limited employment opportunities that appear to inhibit forms of empowerment such as women’s mobility and presence in the public sphere. Work opportunities for women outside the home, in contrast to home-based economic activities, appear to “speed up” the empowerment process. Employment opportunities for men are important both because they help reduce household poverty and because they make men less apt to resent women’s employment. Exposure to middle class, educated lifestyles, nonviolent models of masculinity, and mutuality in spousal relationships appear to have a powerful effect in countering the patriarchal gender norms that support IPV as women become empowered. Particularly in more isolated communities with less direct exposure to these alternative norms, television may be an effective medium for instilling aspirations and promoting lifestyles that exclude IPV (assuming, of course, access to electricity). In summary, our findings suggest that there is much to be learned from existing variations in IPV prevalence among communities and other population subgroups that could inform interventions to mitigate the increased risk of IPV that is often associated with gender transitions.

Footnotes

Acknowledgements

The authors thank Dr. Sultan Hafeez Rahman, executive director, and Dr. Minhaj Mahmud, director of research of the BRAC Institute for Governance & Development (BIGD), BRAC University, and research team members Khurshida Begum, Shefali Akter, Mossabber Hossain, and Dr. Sohela Nazneen.

Authors’ Note

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health, FHI 360, BRAC Institute for Governance and Development, or Columbia University.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health under award number R01HD061630 to FHI 360 (PI S.R. Schuler).

Author Biographies

Sidney Ruth Schuler, a social anthropologist trained at Harvard University, specializes in qualitative and mixed-methods research contributing to policy dialogue and program design on issues of gender in international health and development. She is known especially for her work in conceptualizing and measuring women’s empowerment, for her field studies documenting how reproductive health policies and programs are perceived by intended beneficiaries, and for her work on child marriage and gender-based violence. When the research was conducted she was a social scientist in the Social and Behavioral Health Sciences Division at FHI 360.

Rachel Lenzi, MPH, is a research associate in the Social and Behavioral Health Sciences Division at FHI 360. She specializes in using applied research methods to design and evaluate behavior change interventions. Much of her work focuses on the use of social and behavior change communication and integrated health approaches to accelerate uptake of healthy behaviors, as well as the potential of mobile technology to improve both programming and monitoring and evaluation. Her work spans multiple health sectors and multiple countries in South Asia and sub-Saharan Africa.

Shamsul Huda Badal is a Bangladeshi researcher who has been involved with qualitative research in rural Bangladesh since 1991. He has been involved with numerous studies on the relationship between women’s empowerment and intimate partner violence (IPV) in various capacities, including data collection, interpretation, and analysis, as well as management and coordination of research.

Lisa M. Bates, SM, SD, is assistant professor, Department of Epidemiology at the Mailman School of Public Health, Columbia University in New York. Her research focuses on social determinants of health, with a particular emphasis on processes of social change. She has been engaged in mixed-methods research in Bangladesh for over a decade investigating the relationships between women’s empowerment, gender inequality, marital dynamics, and IPV and other health outcomes.

References

Ackerson

L. K.

Kawachi

Barbeau

E. M.

Subramanian

S. V.

(2008). Effects of individual and proximate educational context on intimate partner violence: A population-based study of women in India. American Journal of Public Health, 98, 507-514.

Amin

Diamond

Naved

R. T.

Newby

(1998). Transitions to adulthood of female garment-factory workers in Bangladesh. Studies in Family Planning, 29, 185-200.

Amin

Khan

Rahman

Naved

R. T.

(2013). Mapping violence against women in Bangladesh: A multilevel analysis of Demographic and Health Survey data. Dhaka, Bangladesh.

Anderson

(1997). Gender, status, and domestic violence: An integration of feminist and family violence approaches. Journal of Marriage and Family, 59, 655-669.

Atkinson

M. P.

Greenstein

T. N. G.

Lang

M. M.

(2005). For women, breadwinning can be dangerous: Gendered resource theory and wife abuse. Journal of Marriage and Family, 67, 1137-1148.

Bates

Osypuk

Naved

R. T.

Yount

Schuler

S. R.

(2015, June). Spousal education gaps and intimate partner violence in rural Bangladesh. Paper presented at the Society for Epidemiologic Research Annual Meeting, Denver, CO.

Bates

L. M.

Schuler

S. R.

Islam

(2004). Socioeconomic factors and processes associated with domestic violence in rural Bangladesh. International Family Planning Perspectives, 30, 190-199. doi:10.1363/ifpp.30.139.04

Cain

Khanam

Nahar

(1979). Class, patriarchy, and the structure of women’s work in rural Bangladesh. Population and Development Review, 5, 405-438.

Campbell

Jones

A. S.

Dienemann

Kub

Schollenberger

O’Campo

. (2002). Intimate partner violence and physical health consequences. Archives of Internal Medicine, 162, 1157-1163.

10.

Choi

Ting

(2008). Wife beating in South Africa: An imbalance theory of resources and power. Journal of Interpersonal Violence, 23, 834-852.

11.

Counts

Brown

J. T.

Campbell

(1992). Sanctions and sanctuary: Cultural analysis of the beating of wives. Boulder, CO: Westview Press.

12.

Dalal

(2011). Does economic empowerment protect women from intimate partner violence? Journal of Injury & Violence Research, 3, 35-44. doi:10.5249/jivr.v3i1.76

13.

Devries

Mak

Garcia-Moreno

(2013). The global prevalence of intimate partner violence against women. Science, 340, 1527-1528.

14.

Dobash

(1979). Violence against wives. New York: Free Press.

15.

Hadi

(2000). Prevalence and correlates of the risk of marital sexual violence in Bangladesh. Journal of Interpersonal Violence, 15, 787-805.

16.

Hashemi

S. M.

Schuler

S. R.

Riley

A. P.

(1996). Rural credit programs and women’s empowerment in Bangladesh. World Development, 24, 635-653.

17.

Heise

(1998). Violence against women: An integrated, ecological framework. Violence Against Women, 4, 262-290.

18.

Heise

(2011). What works to prevent partner violence? An evidence overview. Working Paper, STRIVE Research Consortium. London, UK: London School of Hygiene and Tropical Medicine.

19.

Jahan

(2012). Women’s economic empowerment and violence against women: Exploring the relationship. Washington, DC: World Bank.

20.

Jaim

Mahabub

(2011). Women’s participation in agriculture in Bangladesh 1988-2008: Changes and determinants. Paper presented at the 7th Asian Society of Agricultural Economists International Conference, 13-15 October 2011. Hanoi, Vietnam

21.

Jewkes

Levin

Penn-Kekana

(2002). Risk factors for domestic violence: Findings from a South African cross-sectional study. Social Science & Medicine, 55, 1603-1617.

22.

Kabeer

(2001). The power to choose: Bangladeshi women and labour market decisions in London and Dhaka. London, England: Verso Books.

23.

Kalmuss

Strauss

(1982). Wife’s marital dependency and wife abuse. Journal of Marriage and Family, 44, 277-286.

24.

Kishor

Johnson

(2004). Profiling domestic violence: A multi-country study. Calverton, MD: ORC MACRO.

25.

Koenig

M. A.

Ahmed

Hossain

M. B.

Khorshed Alam Mozumder

A. B.

(2003). Women’s status and domestic violence in rural Bangladesh: Individual- and community-level effects. Demography, 40, 269-288.

26.

Macmillan

Gartner

(1999). When she brings home the bacon: Labor-force participation and the risk of spousal violence against women. Journal of Marriage and Family, 61, 947-958.

27.

Malhotra

Schuler

S. R.

(2005). Women’s empowerment as a variable in international development. In Naranyan

(Ed.), Measuring empowerment: Cross-disciplinary perspectives (pp. 71-88). Washington, DC: World Bank.

28.

Mannan

M. A.

(2004). Baseline survey for assessing attitudes and practices of male and female members and in-laws towards gender based violence. Dhaka, Bangladesh: Bangladesh Institute of Development Studies.

29.

Naved

Persson

L. A.

(2005). Factors associated with spousal physical violence against women in Bangladesh. Studies in Family Planning, 36, 289-300.

30.

Rahman

Hoque

M. A.

Makinoda

(2011). Intimate partner violence against women: Is women empowerment a reducing factor? A study from a national Bangladeshi sample. Journal of Family Violence, 26, 411-420.

31.

Schuler

S. R.

(2007). Rural Bangladesh: Sound policies, evolving gender norms and family strategies. In Lewis

M. A.

Lockheed

M. E.

(Eds.), Exclusion, gender and education: Case studies from the developing world (pp. 179-203). Washington, DC: Center for Global Development.

32.

Schuler

S. R.

Hashemi

S. M.

(1994). Credit programs, women’s empowerment, and contraceptive use in rural Bangladesh. Studies in Family Planning, 25, 65-76.

33.

Schuler

S. R.

Hashemi

S. M.

Badal

S. H.

(1998). Men’s violence against women in rural Bangladesh: Undermined or exacerbated by microcredit programmes? Development in Practice, 8, 148-157. doi:10.1080/09614529853774

34.

Schuler

S. R.

Hashemi

S. M.

Riley

A. P.

Akhter

(1996). Credit programs, patriarchy and men’s violence against women in rural Bangladesh. Social Science & Medicine, 43, 1729-1742.

35.

Schuler

S. R.

Lenzi

Nazneen

Bates

L. M.

(2013). Perceived decline in intimate partner violence against women in Bangladesh: Qualitative evidence. Studies in Family Planning, 44, 243-257.

36.

Schuler

S. R.

Nazneen

Lemons

. (Unpublished results). Changing dimensions of women’s empowerment and intimate partner violence: Perspectives of Bangladeshi women.

37.

Solotaroff

J. L.

Pande

R. P.

(2014). Violence against women and girls: Lessons from South Asia. Washington, DC: The World Bank.

38.

Vyas

Watts

(2009). How does economic empowerment affect women’s risk of intimate partner violence in low and middle income countries? A systematic review of published evidence. Journal of International Development, 21, 577-602.

39.

World Bank. (2003). The Bangladesh primary education stipend project: A descriptive analysis. Washington, DC: Author.

40.

World Bank. (2006). Bangladesh country assistance strategy 2006-2009. Dhaka, Bangladesh: Author.