## Martha C. Nussbaum and Jonathan Glover

Print publication date: 1995

Print ISBN-13: 9780198289647

Published to Oxford Scholarship Online: November 2003

DOI: 10.1093/0198289642.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 18 June 2021

# A Matter of Survival: Women's Right to Employment in India and Bangladesh

Chapter:
(p.37) A Matter of Survival: Women's Right to Employment in India and Bangladesh
Source:
Women, Culture, and Development
Publisher:
Oxford University Press
DOI:10.1093/0198289642.003.0002

# Abstract and Keywords

Chen's paper focuses on women's right to work in rural Bangladesh and India to illustrate the symbolism of independence and the practical necessity for women in the developing world to break from the constraints of custom to forge their own way to economic security. Chen's fieldwork in Bangladesh shows that local traditions and local policies can change in response to the contingencies of human development, while her data in India supports the conclusion that local tradition varies across castes, regions, and time. Chen ends with a four‐dimensional analysis of women's right to gainful employment as a matter of (1) immediate survival for women and their dependents; (2) female mortality rates; (3) women's status; and (4) human justice.

I worked in the fields at night, by moonlight, or at times when there was the least likelihood of being seen. I did any kind of work I could find.

I may die, but still I cannot go out. If there's something in the house, we eat. Otherwise, we go to sleep.

Metha Bai, India (1991)

November 1975: a food‐for‐work site 40 kilometres west of Dhaka city, Bangladesh. Bangladesh is still recovering from a famine in 1974. Saleha Begum, the leader of a team of female labourers, tells the story of how she first began to work outside her home after she and her husband became landless. Despite family and community opposition, Saleha decided to work and prove how hard she could work. Initially she was embarrassed to be seen working, so she often worked at night or at other times when she was not likely to be seen. As she grew accustomed to the work, the criticism also died down. However, at the time when we found Saleha sitting in the shade of a tree near a food‐for‐work site, she and the other women had just been refused work at that site. The local officials turned them back, saying: ‘Women in Bangladesh should not work outside their homes. We have never hired women at food‐for‐work sites.’

February 1991: a village 40 kilometres north of Udaipur city, Rajasthan, India. Metha Bai, an upper‐caste widow with two minor sons, describes her plight as a young widow from a caste which prohibits women from working outside the homestead. Before her husband's death two years before, Metha Bai had not been allowed to work outside their homestead, even to fetch water or fuelwood (her husband performed all outside chores). Now, she helps her father cultivate the land she inherited from her husband and fetches water and fuelwood. However, her in‐laws do not allow Metha Bai to engage in any outside gainful employment. Her only source of support is her father, who helps her till the land and brings regular gifts of food and clothing.

This paper explores the predicament of poor women in poor economies, like Saleha Begum and Metha Bai, who must break with tradition and act independently because they lack the security the tradition is supposed to offer. In communities where women are secluded, perhaps the most conspicuous and yet necessary way for women to break with tradition is to leave their courtyards or homesteads in search of work.

(p.38) The constitutions of Bangladesh and India guarantee women equal employment opportunities (as men) and equal pay for equal work. However, there is a remarkable degree of gender inequality in work opportunities and remuneration, and a remarkable range of variation in female labour force participation in both countries. Women who need or want to engage in gainful employment face one or more of the following constraints: norms of seclusion which confine women to their homes, segmentation of wage labour markets by gender, or exploitation within given trades or industries. This paper focuses primarily on the first type of constraint; the system of secluding women which denies them the right to gainful employment outside the home.

Most communities in the belt of South Asia which extends from Pakistan to Bangladesh across North India share a basic patriarchal kinship system, of which a sexual division of labour and status‐based norms of female behaviour and seclusion are common elements. Admittedly, this system does not operate in exactly the same way for different religions, castes, or communities. Indeed, this paper will explore some differences across caste, class, and region within India and Bangladesh. However, the shared patriarchal ideology of the region has common implications for the status of women in general and for the occupational options of women in particular.

One of the prominent elements of this basic patriarchal system is a division of labour by gender. Under this sexual division, certain types of work are designated as male or female. In both India and Bangladesh, for example, ploughing is almost exclusively a male task whereas drying and storing grain are typically female tasks. Moreover, certain spheres of economic activity are designated as male or female. The homestead or private sphere is predominantly female; the public sphere of markets, roads, and towns is predominantly male. The intermediate sphere of fields and villages is both male and female in India but remains predominantly male in Bangladesh, where female seclusion is more widely enforced.

This shared patriarchal system is interwoven with a hierarchical social structure which, by ranking work appropriate to the status of each caste, further determines patterns of female work. An important symbol of a household's position in the social‐status hierarchy is the type of work its women are allowed to do. Indeed, there is a sharp and systematic decline in the participation of women in work outside the home as wealth and status increase. So much so that, in both Bangladesh and North India, rural households can be divided into four main status groups according to the kind of work women are allowed to do.1 In the top group of high‐caste or (p.39) surplus farm households, in order to preserve household status, women take little or no part in any outdoor activities and often remain secluded in their homes. Below this group are the main local peasant caste or middle farmer group. Their women are occupied mainly with domestic duties but also work in their own fields. In the third group of middle castes or small farm households, women work mainly in their own fields but might work for a wage in the busy season. The fourth and lowest social group is composed of women belonging to the lowest castes or poorest households who regularly seek paid work in order to support their families (Boserup, 1970; Chen, 1983).

In the first part of this paper, I describe how women broke with tradition to join the work force in the aftermath of the 1974 famine of Bangladesh, and discuss the public response to their act of defiance. The story illustrates how local traditions and local policies can evolve and change in response to the contingencies of real life; in this case, the increasing demand for employment by large numbers of women in poor and female‐headed households. It also illustrates a range of perspectives on women's right to employment both within the country and within the international donor community. In the second part of the paper, I describe the predicament of female heads of households from caste groups in India which, as a means of retaining social status or aspiring to ‘higher’ status, prohibits women from seeking gainful employment outside the home. The discussion illustrates how local tradition not only varies across castes and regions but also changes over time. In the final concluding section, I discuss the issue of women's right to gainful employment: first, as a matter of immediate survival for individual women and their families; secondly, as a matter, indirectly, of female mortality rates; thirdly, as a matter of women's status; and, finally, as a matter of human justice.

On a personal note, let me explain my background as a non‐philosopher. I have worked for 20 years in India and Bangladesh. The focus of my concerns has been the economic status of poor women. I have sought through dialogue and action with poor women to identify the constraints to their full economic participation and productivity. In Bangladesh, I worked for the Bangladesh Rural Advancement Committee (BRAC), an indigenous private development agency. We developed a programme for women who, like Saleha Begum, needed to break out of enforced seclusion to enter the labour force. By providing training, credit, and extension services, we were able to enhance the productivity of women's traditional home‐based work and to expand the range of women's activities outside their homes. My subsequent work with Oxfam America in India focused on women working as wage labourers, as home‐based workers, and as vendors. In partnership with over 50 private development agencies, we developed a programme in support of women engaged in the urban informal sector and in several rural sectors. My current research on the economic status of (p.40) widows in rural India has led me back full circle to the problems of secluded women, like Metha Bai, who must seek gainful employment outside their homes in order to provide for their children.

Martha Nussbaum (1993) characterized BRAC's style of operation as ‘a combination of Aristotelian commitment to the human good with Aristotelian contextual sensitivity’. I would add that many of the indigenous private development agencies with which I have had the privilege to work share this style of operation. Committed to one or more human goods, such as women's right to gainful employment, they enter a participatory dialogue with local communities to seek appropriate and acceptable ways to promote these goods. Many are committed to women's right to employment because they have learned that most women from poor households need to earn in order, simply, to survive.

Until the recent past, when landlessness, poverty, and famine forced women to enter the labour markets, the sexual division of labour was more widely and uniformly enforced in Bangladesh than elsewhere in the region. Traditionally, most women in Bangladesh were physically confined to their homesteads or, if they needed to seek wage work, to the homesteads of others in their village. Most women did not work in the fields, even their own, or at construction sites; they did not go to markets, either to buy or sell. Women were effectively excluded from the public ‘male’ sphere of fields, markets, roads, and towns, and essentially confined to the private ‘female’ sphere of huts and homesteads. They were permitted to move about only at prescribed times and for prescribed purposes.

Purdah, which means, literally, ‘curtain’ or ‘veil’, is used figuratively to designate what is proper demeanour or behaviour for women, as opposed to what is improper (bepurdah). In its strictest form, purdah involves the seclusion of women within the boundaries of their homes and the veiling of women outside their homes. In a broader sense, purdah is used to designate what is ‘appropriate’ or ‘respectable’ work for women; what has been called ‘occupational purdah’ (Bardhan, 1985). The following story tells how women from poor and female‐headed households in Bangladesh defied this occupational purdah to seek work.

From the end of June to early September 1974, Bangladesh suffered severe floods. Reports of starvation could be heard immediately following the flood, and grew in severity. The Government of Bangladesh officially declared famine in late September (Sen, 1981). From early October to late November, the government operated langarkhanas (gruel kitchens or free cooked‐food distribution centres) to provide free food to destitute people (p.41) (Alamgir, 1980; Sen, 1981). Migrants from rural areas flooded the streets of Dhaka and other cities in search of food. Although by late November the famine was officially over, the effects of the famine dragged on.

One of the famine districts was Mymensingh in the north.2 In late 1974, UNICEF staff visited the area and observed large numbers of women begging in Jamalpur town. In response, UNICEF decided to introduce an experimental food‐for‐work programme for women outside the town. They expected 100 women to participate: 840 joined the programme. The Jamalpur experience was symptomatic of a widespread problem: during and after the famine of 1974, large numbers of women had joined the ranks of those seeking wage labour opportunities. As late as November 1975, when the Government of Bangladesh (with international donor support) introduced food‐for‐work schemes throughout the country, women came forward in record numbers.

Bangladesh has a long history of food‐for‐work programmes in lean employment seasons and after floods or other crises, but the participation of women was a new phenomenon. The government was simply not prepared to accept women into its rural works programmes. Conditioned by tradition to believe women should not seek outside gainful employment, local officials regularly turned women back. Where they were recruited, women faced problems related to the work and payment norms.

In 1975, prompted by what we had seen and heard in Jamalpur, a group of women (of which I was one) began informally to meet and interview women at food‐for‐work sites in different parts of the country. That was when we met Saleha Begum. We found that the women faced two main problems: initially, many had been denied work; eventually, if they were employed, they were paid less than men. Because payments are tied to the amount of earth moved and because women in Bangladesh had no background in earth‐work, women labourers consistently earned less than male labourers. We began to discuss these problems with responsible government and donor officials.

In March 1976 the Ministry of Relief and Rehabilitation requested advice on the integration of women into its food‐for‐work activities. The United Nation's World Food Programme formalized our ‘ad hoc’ group into an Advisory Committee on Women's Participation in Food‐for‐Work. The Committee's mandate was ‘to advise an appropriate and effective means fully to integrate women into the food‐for‐work activities’ (World Food Programme, 1977:1). After a number of field visits and numerous interviews with women labourers, the Committee, in a meeting with President Zia‐ur‐Rahman and his Cabinet, made two key recommendations: (p.42) that women be recruited in equal numbers as men and be paid the same as men (by adopting a separate work‐payment norm for women). The President endorsed the recommendations and agreed to enforce them.

At a subsequent meeting with the donors involved in food‐for‐work programmes, there was a mixed reaction. Some donor representatives endorsed the recommendations and agreed to the specific policy measures required to promote women's equal participation and payment. Others were resistant to the recommendation of a separate work norm, arguing that women workers were generally supplementary, not primary, earners; that women should be able to lift and carry as much earth as men in a given time period; and that separate norms would encourage false entries in the work registers (i.e., that officials would list men workers as women workers in order to be allotted more wheat for specific schemes).

In response to these objections, the Advisory Committee asked two of us on the Committee to undertake a research study designed to determine the marital status, dependency ratio, work output, and energy requirements of women workers at food‐for‐work sites. With a team of investigators, we interviewed 303 women at 11 food‐for‐work sites in nine districts of Bangladesh. Fifty of this sample were selected for detailed life histories.

Our research revealed the following characteristics of women labourers (Chen and Ghuznavi, 1977). Women of all age groups worked at the sites. The vast majority were between 20 and 50 years of age. One‐third were married. The husbands of most married women were either unable to earn or did not earn enough to meet family needs. Another third were widowed. The sons of most widows either could not be depended on as a source of support or did not earn enough to meet their own and their mother's needs. The remaining third were unmarried, deserted, or divorced. All came from households with insufficient male earnings to meet subsistence needs. Just under half (47%) were the principal income earners for their families. They averaged 3.7 dependants each. In brief, all the women, whether sole, primary, or supplementary earners in their families, worked because they had to.

One of the women we interviewed, a married woman with a disabled husband, evoked the frightening predicament of women forced to provide for their families when male support is withdrawn or proves insufficient:

One day, five years ago my husband returned home with a high fever, which continued unabated for six days, in spite of medication. After that he became bloated and doctors could not do anything to ease the condition. For the first time in my life, I knew what real fear was. I did all I could to get him the best medical care. I looked after him day and night but he did not get better. Today, he is still an invalid.

I was overwhelmed by my problem; I sold all our possessions to get my husband cured but it was not to be. We were faced with an impossible situation. My children were always hungry, there was never enough for them or for me and my husband. (p.43) I tried to get help from my brothers but by then no one's financial condition allowed them to offer any real help and I needed far too much.

So I decided that I would take the only recourse left to me. I would find work and support my family. And so I did. I accepted any kind of work I could find just so long as it helped us to survive. I have not been able to send my children to schools as before but there was nothing I could do about that. Today, I deeply regret not having learnt to weave from my husband.3 I could have had a profession to see me through these difficult days—without always dreading whether I will have work tomorrow or whether my children will be crying in hunger?

When I first heard of earth‐work I was hesitant about going for it. I was not sure I would be able to do those long hours or arduous work. Look at me today. I can not only do it but do it well. I earn enough wheat for my family's consumption. I even manage to save a little. It is not work that I am afraid of but the lack of work. (Chen and Ghuznavi, 1980: 159–60)

The work output findings indicated that women, on average, moved 40 cubic feet of earth per day whereas men, on average, moved 70 cubic feet (Chen and Ghuznavi, 1977). To move these respective amounts of earth, taking average body weights into account, it was estimated that women would require 2,100 calories and men 3,000. A payment schedule for both sexes based only upon the male work norm would result in women receiving little more than half (57%) the wheat payment that men receive. Yet women's nutritional requirement would be over two‐thirds (70%) that of men.

The research findings served to legitimize our earlier policy recommendations. Since 1976, the Government of Bangladesh and the responsible donors have promoted women's participation in food‐for‐work by recruiting female labourers more actively, by reserving separate areas or tasks at all sites for women, by operating all‐women sites, and by adopting separate work‐payment norms for women. In addition, special schemes for women, notably roadside tree‐planting and road maintenance schemes, have been introduced.

Saleha's story has a happy ending. In early 1976, Saleha Begum and her co‐workers were hired at a local food‐for‐work scheme. And, later that year, they were organized into a local women's group by BRAC. In every village where it works, BRAC organizes the poor into separate men's and women's groups. These groups not only are the beneficiaries of BRAC's programmes (in credit, enterprise development, health, and education) but also help shape the content and direction of these programmes and provide village‐level cadre to help implement these programmes.

In 1977, Saleha's group, the Pachbarol Working Women's Group, collectively husked paddy for sale and cultivated potatoes and sugar cane on leased land. They leased four acres of land and purchased paddy seedlings (p.44) to transplant. They contracted two male relatives (including Saleha's husband) as their ‘technical experts’. These men taught the women to broadcast and transplant paddy. When asked to describe the activities of her group, Saleha became animated: ‘If we can handle livestock, surely we can transplant seedlings. Next year we will be able to perform all the agricultural tasks—including ploughing.’

Under her leadership, the women in Saleha's village took advantage of all that BRAC had to offer: they attended non‐formal education classes; they negotiated loans; they received training in poultry rearing, fish culture, silkworm rearing, and agriculture; and they selected women from their group to be trained as paramedics, paraveterinarians, and paralegals. Eventually, Saleha Begum was elected to represent her women's group in the BRAC‐organized local federation of women's and men's groups. A few years later, the federation nominated her as a candidate for local political office. Although she lost the local election, the fact that she did so by only a narrow margin was celebrated as a near‐victory by the federation.

In retrospect, Saleha's victory at the food‐for‐work site was as significant as her near‐victory in the local election. Saleha, her co‐workers and the other women who demanded to be employed at food‐for‐work sites paved the way for countless other women to enter the labour force. They made ‘visible’ the plight of women in households where male incomes are insufficient to meet subsistence needs. They alerted national and international policy‐makers to the economic contributions of women in such households and to the urgent need of such women for gainful employment.

Policy‐makers in both the government and non‐government sectors responded to this need. The government promoted large‐scale wage employment of women through food‐for‐work programmes. By the late 1980s, over 95,000 women were being offered employment for more than 8.5 million womandays each year in two major food‐for‐work programmes. Two private development agencies, the Grameen Bank and BRAC, promoted the self‐employment of women through rural credit. Between them, the Grameen Bank and BRAC have provided loans to over two million women. The women loanees have proved to be not only responsible loanees but also effective entrepreneurs, managing the day‐to‐day subsistence needs of their families.

Official estimates indicate that the female labour force rose threefold from 0.9 to 2.7 million between 1961 and 1985. A 1989 United Nation's study found that about 60% of landless and virtually 100% of female‐headed households report female income earning activities either in wage work or self‐employment (Safilios‐Rothschild and Mahmud, 1989). Changes in the definition used to measure the size of the labour force and the participation of females account for some of this increase. However, the combined effect of increasing population, landlessness, and poverty and of (p.45) expanded work opportunities for women account for most of this increase. Some credit for the increased work opportunities should go to the women, like Saleha, who first defied tradition and entered the wage labour market, thereby paving the way for others.

# 2 Bound by Tradition: The Plight of High‐Caste Women in Rural North India

As noted above, most communities in North India share a common patriarchal kinship system, which has direct implications for the condition of women in general and for the occupational options of women in particular. The major differences within this basic system are mediated by caste. These differences are particularly pronounced between the two poles of the caste hierarchy: at one end, the highly ‘Sanskritized’ upper castes; and, at the other, the so‐called ‘Untouchables’ or lowest castes.

At the all‐India level, the Hindu hierarchy of castes is traditionally divided into four groups or varna: Brahmin, Kshatriya, Vaishya, and Sudra. The first three varna are considered ‘twice‐born’ or ‘upper’ castes; the fourth varna encompasses a large number of ‘middle’ castes. A fifth group of castes is regarded as ‘untouchable’ and outside the system. These untouchable or outcaste communities are commonly referred to as Harijans, the name given them by Mahatma Gandhi which means, literally, ‘the children of God’.

At the local level, the reality is far more complex. Within any given area, there are innumerable, small, local caste groups called jatis. Mutual rank between these groups is vague, arguable, and alterable (Srinivas et al., 1959: 138). For the purposes of this discussion, it is important to note that the caste system varies from region to region and that there is mobility within the system. That is, individual local castes can, and do, move up and down the social status hierarchy.

The social restrictions on the lifestyles of women tend to become more rigid as one moves up in the caste hierachy (Drèze, 1990: 52). Generally there is more seclusion of females in North India than in the South, and among upper castes and classes than among lower castes and classes (Miller, 1981: 780). Within upper caste communities in North India, women are often secluded and denied the right to gainful employment outside their homes. By contrast, lower caste women have greater freedom to take up gainful employment. Among the innumerable small local castes which constitute the vast ‘middle’ of the caste hierarchy, what is considered appropriate behaviour or work for women is closely linked with the family's position (ascribed or aspired) in the social‐status hierarchy (Bardhan, 1985: 2207). As a means to acquiring status, those who can, follow upper caste norms in regard to women's lifestyles.

(p.46) The striving for status differentiation is particularly strong in respect of the kind of work women are allowed to do.4 So much so, that caste norms in regard to women's work can be taken as a symbol of caste rank. For instance, status‐aspiring castes strive to have their women withdrawn from publicly visible manual labour, which is associated with low caste and tribal women (Bardhan, 1985: 2210). Given the preoccupation with status‐appropriate work for women, villagers themselves often rank castes by whether they allow women to work:

• only at their own farms,
• only within the courtyards/homesteads of others,
• only at the farms of others,
• in other activities within their village,
• in other activities outside their village.

Under this ranking scale, the more secluded the woman the higher her household's status or prestige.

In contemporary rural Bangladesh, as I noted above, the direction of change for women is dictated largely by the increasing rates of landlessness and poverty. So that, despite fairly uniform traditional norms against women's outside gainful employment, more and more women join the labour force each year. In contemporary rural North India, there is no single direction of change, especially among the innumerable small local caste groups in the middle regions of the caste hierarchy. In general, the highest status castes still confine women to their courtyards or homesteads; whereas the lowest status castes allow their women to engage in all types of wage work, including migrant wage work.5 But among the middle caste groups, the norms regarding women's work are more fluid. In those households which suffer impoverishment or loss of male earnings, women often face forced entry into the labour markets. In those households which enjoy upward mobility, women often face forced withdrawal from the labour markets.

While the percentage of the population living in poverty has not decreased significantly, the numbers of households enjoying increased wealth has increased. As households acquire wealth or otherwise aspire to status, given rural India's preoccupation with caste and status, they often try to distance themselves from households perceived to be lower in social status and to imitate higher‐status households. Different terms have been used to (p.47) denote this imitation of upper castes by lower castes. The most common term is ‘Sanskritization’, as certain Vedic or Sanskritic rites are confined to upper castes. The other common terms are ‘Brahminization’ (as Brahmins are most often imitated) and ‘Rajputization’ (as Rajput kings are also imitated).6 The point here is not to argue for one or another term but to note the ‘strength of the tendency to imitate and also the main direction of the tendency’ (Dumont, 1980: 192).

Most social scientists have discussed this process in terms of how households imitate upper caste rituals and ceremonies in an attempt to achieve ritual distance from lower households. A few discuss the process in terms of how households imitate upper caste norms of marriage and norms regarding women: ‘following the orthodox Brahmanical standards in child marriage, in the prohibition of widow remarriage, and in placing their women in seclusion’ (Mandelbaum, 1948: 138). For women, this process of imitation has very specific implications: in particular, for the kind of work women are allowed to do. Generally, this process involves the withdrawal of women from the labour market.

But other processes are also at work in rural India. Not all households are able to imitate upper caste norms. Some are becoming poorer; others have remained poor. Current estimates vary, but as many as 42 per cent of households in rural India may have incomes below the poverty line (World Bank, 1989). And some households face the loss of male support. Again, current estimates vary. Census estimates using official definitions indicate that just under 10% of households in rural India are de jure headed by women (Visaria and Visaria, 1985). As Drèze points out, other studies, using more inclusive definitions, suggest that as many as 30 per cent of Indian households are de facto headed by women (Drèze, 1990).

A large number of women, especially female heads of households, are caught between the contradictory trends of Sanskritization, whereby caste norms confine them to their homesteads, and impoverishment or vulnerability, whereby they need to work in order to provide for their families. These include women from upper caste or status‐aspiring but poor households. Particularly disadvantaged are women from upper caste or status‐aspiring households with insufficient male support to meet subsistence needs, especially those who have suffered a recent sudden loss of male support. The plight of such women is acute because if they enter the labour (p.48) force they risk scorn, censure, and (sometimes) excommunication by their kin and caste groups, but if they do not enter the labour force they risk the welfare of their families.

The story of Metha Bai illustrates this predicament. Metha Bai was widowed two years before I met her, when she was 28. For three years prior to his death, her husband had been bed‐ridden. Earlier, their two young daughters had died. Metha Bai was left with two young sons and a small parcel of land. Before his death, Metha Bai's husband and his brothers had separated their residences and partitioned their land. His mother lives with his younger brother. Metha Bai continues to live next to her in‐laws but they offer no support or maintenance. Instead, they make her life miserable.

Metha Bai faces two major types of problems. First, caste norms which prohibit her from remarrying and from engaging in outside gainful employment. Before her husband's death, Metha Bai had not been allowed to work outside their homestead, not even to fetch water and fuel: her husband performed all outside chores. Now, she is forced to cultivate her share of land with the help of her father and to fetch water and fuel for herself and her sons. However, she is not allowed to engage in wage work or any gainful activities outside her home. As she expressed her predicament: ‘I may die, but still I cannot go out. If there's something in the house, we eat. Otherwise, we go to sleep.’

Metha's second set of problems revolves around her relationship with her in‐laws. Her mother‐in‐law, also widowed, has a fierce temper and reputation. After being widowed, she had several sexual partners and gave birth to a son out of wedlock. Metha Bai's husband used to quarrel regularly with his mother, as he was critical of her behaviour. Now, according to Metha Bai, her mother‐in‐law beats her and mistreats her sons. Both her mother‐in‐law and brothers‐in‐law have put pressure on Metha Bai to forego her claims on her husband's share of land and to leave the village. They will not let her contract a sharecropper to till her land. Allegedly, they forced Metha Bai to sell some of her jewellery to pay for the death ceremonies of her husband, and have taken some of her land and jewellery.

Metha Bai's only source of support is her father, now quite elderly. Her father helps her cultivate her share of land, using his own bullocks and plough which he must being from his village as needed. He also brings regular gifts of food and clothing, whatever her parents can spare. Metha Bai is their eldest child; she has four younger brothers at home. Her constant refrain throughout my interview with her was: ‘When my father dies, I will die.’

During the interview, Metha Bai was patching torn and used shorts and shirts for her two sons. Each item had been patched repeatedly. She used pulled threads from other used clothes and a large, twisted, rusty needle. Clearly, her situation is pathetic. She cannot afford the bus fare to visit her parents. When one of her sons fell ill and was taken to a hospital in Udaipur by her father, she could not afford the bus fare to visit him. And, clearly, (p.49) Metha Bai is deeply troubled and anxious. Throughout our interview, she kept looking over her shoulder to make sure her in‐laws were not nearby. As she explained: ‘My mind does not settle, I'm always upset. All day and night, for the last two years, I am always upset. I will just die.’ When asked why she didn't return to her parental home to live, she had an immediate answer: ‘My brothers will give me food and clothing, but will they give any rights to land or property to my sons? Here, my sons have some rights to land.’

Like other vulnerable upper caste women before her, Metha Bai will probably have to risk censure, abuse, and even disinheritance by her in‐laws if she seeks work in order to feed her sons. She may begin by collecting and selling grass, fuelwood, cow dung cakes (trying to earn an income from the subsistence activities she has recently undertaken); she may move on to perform domestic services in other households; eventually she may work as an agricultural wage labourer. I have met other single upper caste women who have entered the wage labour market. They did so gradually, only after they recovered from the sudden loss of male support, took stock of their predicament, and then took the bold step of seeking gainful employment outside their homes.

It is important to note that the system of seclusion is being imitated not just by middle caste groups but also by lower castes and tribal groups. As part of their struggle to assert their ethnic identities, many tribal communities in which women have traditionally been free to engage in all types of work now pronounce the seclusion of women as part of their group heritage.7 As part of their effort to raise their status, educated lower caste communities in some areas have begun to change their occupations and to seclude their women. For example, the Chandal caste in West Bengal encourages its men to become educated and seek salaried jobs, and, at least in one area, prohibits its women from working outside the home.8

Chobirani Prodhan is a young Chandal widow in Birbhum District, West Bengal. Jean Drèze interviewed her in 1988, three months after she was widowed:

When I met her in 1988, Chobi Prodhan had only been widowed for three months. She had four young daughters, and no sons. She and her children had subsisted since the death of her husband partly from the meagre produce of the half‐acre of land they possess, and partly from a loan obtained from the local credit cooperative (they own no animals). But the future looked extremely bleak, and Chobi Prodhan was desperately anxious to find a means of surviving. Since her husband's death, she had not taken up wage labour, or any other kind of income‐earning activity. (p.50) She laments, ‘aami kicchu kam jaani na’, ‘I don't know any work’. (Drèze, 1990: 82–3)

I interviewed Chobirani in early 1991, three years after she was widowed. In 1990, she sold her share of a local pond (pukkur) to marry one of her daughters. She is currently rearing a female calf on a sharecrop basis: in return for rearing the calf until its first delivery, she will be given one of its calves. Otherwise, her situation has not changed. For six months she and her remaining three daughters subsist off their share of harvest from their land; they have to take loans to meet their subsistence needs for the other six months of each year. When asked why she hasn't taken up any other economic activity, Chobirani explains that she was raised and educated in a nearby town, that she has few functional village skills, and that the Chandal caste elders discourage women from seeking outside work. Three years after her husband's death, she is still desperately anxious to find a means of survival.

In India, womens' employment has been a subject of growing interest both to social scientists and to policy‐makers. However, the problems faced by a large female labour force have overshadowed the problems faced by secluded women. The public policy debate on women's work has centred on three broad sets of issues. First, the highly segmented labour markets in which women are typically relegated to low paying and insecure jobs. Secondly, exploitative relationships within given trades or industries which prevent women (and men) from realizing the full returns to their labour. And, thirdly, the failure of policy‐makers to recognize women workers as legitimate clients of mainstream development programmes and policies. The patriarchal system of secluding women, which denies some women the right to gainful employment outside the home, has received comparatively less policy attention.

# 3. Women's Right to Employment

## 3.1. A Matter of Justice as Well as Survival

As the preceding discussion has, I hope, illustrated, local traditions in regard to women's work and status are not uniform or unchanging in India and Bangladesh. Within the markedly hierarchical and patriarchal traditions of both countries, there has always been some room for mobility and manœuvring. As we have seen, change has often been initiated from below: by lower class or caste groups enabled by wealth or education to imitate upper caste‐class norms or forced by impoverishment to violate upper caste‐class norms.

It is important to note that change is also initiated and controlled from above: by upper classes and castes; by caste or village elders; by family (p.51) patriarchs. The poor women with whom BRAC worked in Bangladesh often blamed the rich for dictating the norms that prohibited them from seeking work outside the village. Women at a BRAC‐organized workshop described how the rich controlled their options:

Women work whole days in the homes of the rich: boiling paddy, drying paddy, cooking rice for them. In return, they are given only watery rice, boiled wheat or one quarter or half seer rice for a whole day's work.9 They also have no fixed working hours which men have. The women start working early in the morning and until night . . . It is the village leaders' strategy to keep them working at low wages in their homes. If the women work outside, without purdah, then the rich people will not be able to advance credit to poor women at exhorbitant rates of interest. If the women work outside they might be economically better off . . . The rich in their own interest have made norms and laws. To suppress the poor, the rich, the matbors (village elders), and the mullahs (religious leaders) formulate religious policies and impose on them certain religious injunctions. . . (Chen, 1983: 72–3)

In subsequent discussions with BRAC staff, I explored the women's line of analysis. I wanted to know how the rich managed to control the work options for women. What is at stake, the staff explained, is the definition of what is purdah (within the norms of purdah) and what is bepurdah (outside the norms of purdah). The rich and powerful in a village are able to adjust and change the norms of purdah to fit their convenience, as follows:

What is necessary for their wives to do is sanctioned as purdah. For example, if women from rich households need to go to the town to appear in court, even to remain in town for a few days, this is sanctioned as within the norms of purdah. When women from a BRAC‐organized group want to go to town to attend a workshop or meeting, even for a single day, their action is condemned as bepurdah. (Ibid.: 73)

In my recent interviews with widows in rural North India, many widows blamed caste elders and family patriarchs for limiting their options. When asked why they did not work outside their homes, many said they would either be ‘excommunicated’, ‘disowned’ or ‘ostracized’ by their husband's or their own family, kin, and caste. By excommunication, they meant they would be forced to leave the village in which they lived. By disowned, they meant they would be forced to give up their rightful shares to land or other property. By ostracized, they meant they would not be invited to attend any ceremonial, ritual, or social occasions. In contrast, some widows reported that their caste elders or in‐laws, recognizing their need to work, had allowed them to seek work outside their home. Clearly, tradition is a human creation and the interpretation or enforcement of tradition is at some person's or some group's discretion: in many cases, at the discretion of the rich and powerful.

## (p.52) 3.2. A Matter of Immediate Survival

Women's right to gainful employment can be conceptualized and discussed from several perspectives: as a matter of immediate survival for individual women and their families; as a matter, indirectly, of female mortality rates; as a determinant of women's status; and as a matter of human justice. This paper has focused on individual women and the immediate needs of their families. As the paper suggests, women's right to gainful employment in India and Bangladesh is not a casual matter; it is a matter of immediate survival in many households.

There is growing evidence of the centrality of women's work and income in meeting the immediate day‐to‐day requirements of subsistence or survival. In poor households throughout the developing world, the welfare of individual women or their families is directly related to the ways women are allowed to earn an income. And current demographic and social trends suggest that increasing proportions of women throughout the developing world will bear the primary, if not sole, responsibility for the economic welfare of themselves and their children. Clearly, this feminization of poverty calls for policies which promote women's full economic participation and productivity.

## 3.3. A Matter of Female Mortality Rates

In recent years, some social scientists, notably Amartya Sen and Jean Drèze, have drawn attention to the importance of female participation in ‘gainful’ economic activity as a determinant of women's overall socio‐economic status as well as their bargaining position within their families. In their book Hunger and Public Action, Drèze and Sen point out that ‘there is considerable evidence that greater involvement with outside work and paid employment does tend to go with less anti‐female bias in intra‐family distribution’ (Drèze and Sen, 1989: 58). They compare the ratios of ‘economic activity rates’ (roughly pertaining to outside work, including paid employment) of women vis‐à‐vis men, and the ratios of female life expectancy to male life expectancy (Ibid.: 59) (see Table 1). It turns out that the (p.53) ranking of the different regions in terms of life expectancy ratios is almost the same as that in terms of activity‐rate ratios.

Table 1 Activity‐Rate Ratios and Life Expectancy Ratios, 1980

Regions

Activity‐Rate Ratios (female–male)

Life Expectancy Ratios (female–male)

Values

Ranks

Values

Ranks

Non‐Northern Africa

0.645

1

1.071

1

Eastern and Southern Eastern Asia

0.610

2

1.066

2

Western Asia

0.373

3

1.052

3

Southern Asia

0.336

4

0.989

5

Northern Africa

0.158

5

1.050

4

Source: Drèze and Sen (1989: 58).

Within India, differentials in female and male workforce participation rates also seem to correlate with differential mortality rates of female and male infants and children. ‘Across regions and communities in rural India, larger female participation in the generation of income or subsistence production is found to be associated with a more balanced sex‐ratio in the population under age ten’ (Bardhan, 1985:2262). In her analysis of differences in sex ratios between North and South India, Barbara Miller reports a marked difference between North and South India in terms of women's involvement in ‘outside’ work and of juvenile sex ratios. Generally there is a lower female labour force participation rate and a less‐balanced sex ratio in the North than in the South, and among upper castes and classes than among lower castes and classes.10 As Miller concludes, where female labour force participation (FLP) is high ‘there will always be high preservation of female life, but where FLP is low, female children may or may not be preserved’ (Miller, 1981: 117).11

This correlation between the extent to which women work outside the home and the relative life expectancy of females requires further investigation and analysis. But, as Jonathan Glover has noted in his paper for this conference, ‘it seems plausible that, in many communities in developing countries, the discouragement of women's paid work goes with favouring boys over girls in the provision of food and medical care, or with other kinds of female disadvantage’ (Glover, 1991:1). Given the plausibility of the correlation between women's gainful activity rates and women's life expectancy rates, the demand for women's right to work is not simply a demand for female autonomy or independence: it is, in all likelihood, a matter of female survival.

## 3.4. A Matter of Women's Status

It is important to recognize the broader value of gainful employment to women, outside the context of immediate survival. Engel's famous hypothesis that women's emancipation would come about with their full entry into (p.54) social production has been a subject of interest among feminists concerned with development issues, particularly in the Third World (Standing, 1985). Most feminists agree that women's right to gainful employment is a necessary, but not sufficient, precondition for women to achieve personal power and status in the public domain. This is particularly so if increased female labour force participation is accompanied by increased female earnings and by increased female control over income and the means of production.

Gainful employment outside the home, particularly as an independent wage or salaried worker, does, undeniably, affect the self‐image and perceived value of women. It represents a means of escape from male control over female labour (particularly in peasant households). It represents a means of economic independence. It often leads to increased class, gender, and individual consciousness. It often serves to increase women's bargaining power and autonomy within the household. And it generally serves to increase the perceived value of women within the household and within society more broadly.

The right to gainful employment needs, therefore, in the context of this conference to be discussed as an important determinant of female functioning, as a basic female capability. But fundamental change for women cannot be based solely on increased labour force participation. The conditions under which women work must be just. Too many women, particularly in the Third World, work for low wages or under exploitative conditions. Some feminist theorists have identified collective action as a primary step for women in achieving just employment and justice more broadly (Sanday, 1974). Empirical support for this analysis is the impressive record of women's organizations in India which have organized women around economic concerns (Sebstad, 1982; Chen, 1989). Their achievements extend beyond increasing women's earnings and productivity, effecting changes in women's outlooks, or increasing their freedom within the family unit to enabling women to mobilize vital community resources, to gain a voice in local government, and to struggle for a just working environment (Bruce and Dwyer, 1988).

## 3.5. A Matter of Justice

Much has been written about the moral basis of subsistence relations in rural South Asia, particularly in terms of the reciprocal relations between castes in the caste hierarchy (Appadurai, 1984). Less has been written about the moral basis of gender relations within families. However, the literature on the joint family and on kinship relations reports that, under traditional patriarchy, women are to be maintained by their own families before marriage and by their husband's families after marriage. (p.55) And that widows are to be maintained by their husband's families or by grown sons.

Whether or to what degree South Asia's rural economy operated in a moral way in the past is open to question. But there is little question that, given the forces of modernization and the pressures of population growth, there has been an erosion of moral ties within (and between) caste and kin groups. As we have seen in the examples from India and Bangladesh, we can no longer assume that traditional family structures, which relegate women to domesticity and dependence, actually provide women with the supposed social security of that dependence. Indeed, the evidence suggests that increasing numbers of women receive little, if any, social security from traditional family structures and must act independently in order to provide for their families.

This reality poses a moral dilemma. Many women in rural South Asia appear to have been forced out of a situation in which they were maintained while remaining secluded into one in which they must remain at least partially secluded without being securely maintained. The demand that women be allowed to abandon seclusion and seek gainful employment outside the home should not be seen as an outside challenge to local culture and tradition but as a local response to changes in local culture and tradition.

As I hope this paper has shown, even within the shared patriarchal kinship system of Bangladesh and North India, tradition is not uniform or unchanging. There is significant variation across caste and class, even within individual villages, at any given point in time. And there is significant change across time. Tradition is a human creation subject to change. And it seems reasonable for insiders, but also outsiders, to demand change when local traditions put some members of local communities at risk. Given that women's need to earn is a matter of survival for many women and their households, especially in poor economies, it seems reasonable to propose that all women should have the right to gainful and just employment.

As Onora O'Neill 1993 argues, ‘a serious account of justice cannot gloss over the predicaments of impoverished providers in marginalized and developing economies’. In this paper, I have discussed a predicament faced by many impoverished female providers in the Third World: that is, traditional cultural barriers to their equal participation in the workforce. Like Saleha Begum and Metha Bai, many women must break with tradition and act independently because they lack the security that tradition is supposed to offer. Under Martha Nussbaum's formulation of human capabilities (see ‘Human Capabilities, Female Human Beings’, in this volume), the opportunity for gainful employment would be seen as an essential human good, one of the basic human capabilities, without which a basic human life is often not possible. A serious account of justice should not, therefore, gloss over women's right to gainful and just employment.

(p.56) Bibliography

Bibliography references:

Alamgir, Mohiuddin (1980). Famine in South Asia: Political Economy of Mass Starvation. Cambridge, Mass.: Gunn and Hain.

Appadurai, Arjun (1984). ‘How Moral is South Asia's Economy?—A Review Article’, Journal of Asian Studies, XLII, 481–97.

Bardhan, Kalpana (1985). ‘Women's Work, Welfare and Status: Forces of Tradition and Change in India’, Economic and Political Weekly, XX, Dec.

Boserup, Ester (1970). Woman's Role in Economic Development. New York: St. Martin's Press.

Bruce, Judith and Dwyer, D. (1988). ‘Introduction’, in D. Dwyer and J. Bruce (eds.). A Home Divided: Women and Income in the Third World. Stanford: Stanford University Press.

Chen, Martha (1983). A Quiet Revolution: Women in Transition in Rural Bangladesh. Cambridge, Mass.: Schenkman.

—— (1989). ‘The Working Women's Forum: Organizing for Credit and Change in Madras, India’, in Ann Leonard (ed.), SEEDS: Supporting Women's Work in the Third World. New York: Feminist Press.

Chen, Martha and Ghuznavi, R. (1977). Women in Food for Work: The Bangladesh Experience. Rome: World Food Programme.

—— (1980). ‘Women in Bangladesh: Food‐for‐Work and Socio‐Economic Change’, in Alfred de Souza (ed.), Women in Contemporary India and South Asia. New Delhi: Manohar.

Drèze, Jean (1990). Widows in Rural India. DEP Paper No. 26, Development Economics Research Programme. London: London School of Economics.

—— and Sen, A. K. (1989). Hunger and Public Action. Oxford: Clarendon Press.

Dumont, Louis (1980). Homo Hierarchicus: The Caste System and Its Implications, trans. M. Sainsbury, L. Dumont, and B. Gulati. Chicago: University of Chicago Press.

Glover, Jonathan (1991). ‘Introduction to Conference on Human Capabilities: Women, Men, and Equality’. Paper prepared for the UNU/WIDER Conference on Human Capabilities: Women, Men and Equality. Helsinki: WIDER.

Government of India (1951). Census Reports of India: West Bengal. New Delhi: Government of India Publications.

—— (1991). Census of India 1991: Provisional Population Tables. New Delhi: Census Commission.

Mandelbaum, David G. (1948). ‘The Family in India’, Southwestern Journal of Anthropology, 4 (Summer).

Miller, Barbara D. (1981). The Endangered Sex: Neglect of Female Children in Rural North India. Ithaca: Cornell University Press.

Nagaraj, K (1991). ‘The “Missing” Women: The Declining Sex Ratio’, Frontline, 25 May–7 June.

Nussbaum, Martha (1993). ‘Non‐Relative Virtues: An Aristotelian Approach’, in The Quality of Life. Oxford: Clarendon Press.

O'Neill, Onora (1993). ‘Justice, Gender and International Boundaries’, in M. C. Nussbaum and A. Sen (eds.), The Quality of Life. Oxford: Clarendon Press.

(p.57) Risley, H. H. (1981). The Tribes and Castes of Bengal: Ethnographic Glossary, vol. I. Calcutta: Firma Mukhopadhyay.

Safilios‐Rothschild, Constantina and Mahmud, S. (1989). Women's Roles in Agriculture—Present Trends and Potential for Growth. Monograph for Agriculture Sector Review. Dhaka: UNDP and UNIFEM.

Sanday, Peggy R. (1974). ‘Female Status in the Public Domain’, in Michelle Rosaldo and Louise Lamphere (eds.), Women, Culture, and Society. Stanford: Stanford University Press.

Sebstad, Jennefer (1982). Struggle and Development Among Self‐Employed Women: A Report on the Self‐Employed Women's Association, Ahmedabad, India. Office of Urban Development: Bureau for Development Support. Washington, DC: United States Agency for International Development.

Sen, Amartya K. (1981). Poverty and Famines: An Essay on Entitlement and Deprivation. Oxford: Clarendon Press.

—— (1990). ‘Gender and Cooperative Conflicts’, in Irene Tinker (ed.), Persistent Inequalities: Women and World Development. New York: Oxford University Press.

Srinivas, M. N. (1952). Religion and Society Among the Coorgs of South India. Oxford: Oxford University Press.

Srinivas, M. N., Damle, Y. B., Shahani, S., and Beteille, A. (1959). ‘Caste: A Trend Report and Bibliography’, Current Sociology, VIII.

Standing, Hilary (1985). ‘Resources, Wages, and Power: The Impact of Women's Employment on the Urban Bengali Household’, in Haleh Afshar (ed.), Women, Work, and Ideology in the Third World. London: Tavistock Publications.

Visaria, Pravin and Visaria, L. (1985). ‘Indian Households with Female Heads: Their Incidence, Characteristics and Level of Living’, in D. Jain and N. Banerjee (eds.), Tyranny of the Household: Investigative Essays on Women's Work. New Delhi: Shakti Books, Vikas Publishing House.

World Bank (1989). India: Poverty, Employment, and Social Services. A World Bank Country Study. Washington, DC: The World Bank.

World Food Programme (1977). Women in Food for Work: The Bangladesh Experience (Note by the Executive Director). Interim Evaluation and Terminal Reports. Rome: World Food Programme. (p.58)

## Notes:

(1) In predominantly Muslim Bangladesh, where caste is not an explicit social category, households are most usefully ranked in terms of land‐holding classes: surplus, middle, and small farm households plus landless labour households. In predominantly‐Hindu North India, households are usefully ranked in terms of caste. See Boserup (1970: 69–70) for a summary of how caste influences women's work in India, and Chen (1983: 68) for a summary of how the land‐holding class influences women's work in Bangladesh.

(2) The 1974 famine was not due to a decline in food availability (indeed Mymensingh District enjoyed a substantial increase in output in 1974) but to a decline in wage employment and the rice‐entitlement of wages (the floods came at the period of maximum wage employment in normal years). See Sen (1981).

(3) Under the sexual division of labour in weaving communities in Bangladesh, men weave on the loom and women perform ancillary tasks, such as spinning and helping prepare the warp.

(4) Other norms which are linked to the family's position in the social‐status hierarchy include norms of marriage (and remarriage) and norms of female demeanour.

(5) This is not to say that upper caste and lower caste norms regarding women's work are immutable. In some upper caste households in rural India, especially those with an urbanized or ‘Westernized’ education, women are allowed to work at salaried jobs with status. And, a few lower castes in rural India have adopted upper caste norms and seclude their women (as will be discussed below).

(6) The concept of ‘Sanskritization’ can be best presented in the words of M. N. Srinivas who popularized the concept: ‘The caste system is far from a rigid system in which the position of each component caste is fixed for all time. Movement has always been possible, and especially so in the middle regions of the hierarchy. A low caste was able to rise to a higher position in the hierarchy by adopting vegetarianism and teetotalism, and by “Sanskritizing” its ritual and pantheon. In short, it took over as far as possible, the customs, rites and beliefs of the Brahmins, and the adoption of the Brahminical way of life by a low caste seems to have been frequent, though theoretically forbidden. This process has been called “Sanskritization” in this book, in preference to “Brahminization”, as certain Vedic rites are confined to the Brahmins and the two other “twice‐born” castes.’ See Srinivas (1952: 30).

(7) Personal communication, Vina Mazumdar, Centre for Women's Development Studies (June 1991).

(8) Historically, the Chandals were depicted as ‘outcast and helot people, performing menial duties for the Brahmans’ (Risley, 1981: 183). As noted in a glossary of The Tribes and Castes of West Bengal, Census of 1951, among the Chandal community, ‘a considerable number now follow the various so‐called learned professions’ (Government of India, 1951).

(9) A seer is a measure of weight equal approximately to two pounds.

(10) According to the 1981 Census, about 27% of the women in four southern states were workers; the corresponding percentage in the Hindi‐belt in the North was about 16%. See N$a ǎ$garaj (1991). According to the same Census, the sex ratio in the four southern states averaged 987 females to 1,000 males; the corresponding ratio in the Hindi‐belt in the North was 923 females to 1,000 males (Government of India, 1991).

(11) For the region as a whole, the female‐to‐male sex ratios are some of the lowest in the world:

 Pakistan 905 women to 1,000 males in 1986 India 905 women to 1,000 males in 1991 Bangladesh 905 women to 1,000 males in 1986

Sources: For Pakistan and Bangladesh, Drèze and Sen (1989). For India, Government of India (1991).