By Jan Marci Brunson B.A., Eckerd College, 1999 A.M., Brown

REPRODUCING HIERARCHY:

WOMEN’S POSITIONS AND EMBODIMENT OF SOCIAL CHANGE IN THE KATHMANDU

VALLEY

Jan Marci Brunson

B.A., Eckerd College, 1999

A.M., Brown University, 2001

A dissertation submitted in partial fulfillment of the

requirements for the degree of Doctor of Philosophy

in the Department of Anthropology at Brown University

PROVIDENCE, RHODE ISLAND

May 2008

This dissertation by Jan Marci Brunson

is accepted in its present form by the Department of Anthropology

as satisfying the dissertation requirement for

the degree of Doctor of Philosophy.

Date ______Lina M. Fruzzetti, Director

Recommended to the Graduate Council

Date ______David I. Kertzer, Reader

Date ______Daniel J. Smith, Reader

Date ______Lynn Bennett, Reader

Approved by the Graduate Council

Date ______Sheila Bonde Dean of the Graduate School

iii

VITA

Jan M. Brunson was born in Tallahassee, Florida, on October 1, 1977. Her first introduction to anthropology was at Rutherford High School as part of the International

Baccalaureate Program’s curriculum. She went on to major in anthropology at Eckerd

College, and had the opportunity to conduct research in Sri Lanka with her faculty mentor Dr. Victoria Baker through a fellowship from the ASIANetwork Freeman

Foundation Student-Faculty Fellows Program. Her passion for teaching and learning was cultivated by participating in the Ford Scholars Program at Eckerd College. After graduating in 1999 with a thesis that received the distinction of honors, she enrolled in the graduate program in anthropology at Brown University. In addition to her graduate education in the Department of Anthropology, Jan was a trainee of the Population Studies and Training Center (PSTC) at Brown. She also spent three summers at Cornell

University for language training in Nepali. This training culminated in dissertation research in Nepal funded by a Fulbright-Hays Doctoral Dissertation Research Award and supplementary funds from the PSTC. She graduated from Brown University in 2008.

ACKNOWLEDGMENTS

Many people have contributed to this project and some of them in such profound ways that I am unable to do them justice here. Foremost, I have benefited from the guidance of several mentors at Brown University. I am indebted to Lina Fruzzetti for serving as my thesis advisor, and moreover for ensuring that I developed both breadth

and depth in the literature on and of South Asia. I was grateful for Lina’s support as the

focus of my project evolved over time. David Kertzer has been a model scholar and

writer, setting an example and coaching me in both these areas. No matter how

demanding his schedule was, he always made time for me and my research. For setting

the highest professional standards and instilling the same in me, I thank him. Dan Smith

exhibited the perfect mix of encouragement and criticism as a mentor. His generosity

with his time and his willingness to seriously and closely engage my writing and

arguments (even in early drafts) was truly impressive. He has been a major influence in

my intellectual development, and he greatly enhanced my dissertation with his

suggestions. I am grateful to Lynn Bennett for serving as the fourth member and outside

reader on my dissertation committee. Her past and ongoing scholarship has and will

continue to influence my work. The deficiencies that remain despite their suggestions are mine alone.

Also at Brown, I would like thank Pat Symonds in the Department of

Anthropology for her advice on professionalization over the years and her graciousness

and warmth as a mentor. Although a list of the ways she has been of assistance is too

long to include, I am extremely grateful to Kathy Grimaldi for masterfully and cheerfully

handling many of the essential logistical and bureaucratic matters related to graduate

school and also for her hugs.

Much gratitude is also felt toward my colleagues and friends who offered

criticism on various incarnations of the manuscript. For much moral support and

patience with reading early drafts, I am deeply indebted to Pilapa Esara, Andrew

Huebner, and Susi Krehbiel Keefe. I owe special thanks to Andrew and his family for

encouragement during the early stages of writing. For support during the intense final months of writing, I am grateful for Benjamin Young. Benjamin provided much intellectual inspiration, editing assistance, and sustenance, and he frequently devoted his energy to helping me clarify my thinking. My work has benefitted substantially from conversations with him and Jason Sears.

In Nepal, I am grateful to all the people who welcomed me into their homes and their lives, especially the Shrestha family. They provided me with the kind of love that only family can give and vouched for me as an upstanding person to others in their community. Manoj Kumar Shrestha played a crucial role in mapping and taking a household census of the community and in the administration of the survey questionnaire.

He spent countless hours sitting with me and translating tapes of interviews together. His assiduousness in his work as a research assistant was unmatched, and he was always

ready with suggestions to solve the myriad of logistical problems. I would also like to

thank Dambar Pariyar for assisting with administering the initial survey. Mina

Manandhar, my lead research assistant, assisted with translations, but her major

contribution was the perceptiveness, tact, and astuteness with which she conducted

interviews with women in the case studies. I was able to depend upon her to handle

delicate relationships with professionalism and care. She also became an amicable and

valued companion for months of walking up and down the mountainside in the process of

conducting interviews. I also thank Geeta Manandhar, who was more my sage than she

was ever my language teacher, and who acted as an older sister and friend.

In terms of institutional support, I am grateful for financial support for the

research from a Fulbright-Hays Doctoral Dissertation Award and my subsequent

relationship with the Fulbright Commission in Nepal and Michael Gill. For

supplementary research funds, I thank the Population Studies and Training Center at

Brown University. My first few trips to Nepal prior to the dissertation research went

more smoothly because of the assistance of the Cornell-Nepal Study Program, and I am

grateful for the lasting connections and friendships that were first initiated by Kathryn

March and Banu Oja.

Finally, for everything they have done to help me achieve my dreams, and

especially for providing me with the space to define those dreams however I choose, I thank my family – Mom, Dad, and Sherri. I find it somehow appropriate that in Nepal

nieces have particularly loving relationships with their uncles, because I have a doting

uncle, Jimmy, whom I thank for taking such an interest in my success. I dedicate this

dissertation to my grandfather, James E. Stockton, in recognition of his unconditional

love and support throughout my graduate school career. I will always admire his

adherence to his values of honesty, loyalty, and persistence and his dedication to family.

vii

TABLE OF CONTENTS

Acknowledgments v

Prelude: A Convergence of Personas in the Hindu Kingdom 1

Introduction: Women, Agency, and Reproduction 12 Introductions Defining Women Toward a Critical Phenomenology of Procreation Women Experiencing and Producing Social Change

1 Methods and Encounters 40 The Research Setting Enter the Anthropologist Overview of Methods Encounters

2 Social Hierarchies in Context 62 Constructs of Caste in South Asia The Distinctiveness of Caste in Nepal Rethinking Women’s Autonomy and Caste

3 From Daughters to Mothers-in-law: Women’s Positionality over the Life Course 88 Life as a Daughter, Labor as a Daughter-in-Law, Uncertainty as a Mother-in-Law Like a Potter’s Wheel: Women and the Family Cycle

4 Describing and Inscribing Reproductive Bodies 116 A Descriptive Overview of Maternal Health Hidden Pregnancies Home Births and Hospital Emergencies Protected Postpartum Periods

5 Conflicting Discourses and Agentive Bodies 140 The Unlikely “Happy Family” Son Preference and Fertility Decline What Good Are Sons? Deliberating over Giving Birth Again

Conclusion: Reproductive Realities in Nepal 167

Bibliography 174

viii

LIST OF FIGURES

Figure 1.1 Ethnic and Caste Groups in Vishnumati, 2001 46 Figure 1.2 Parbatiya Caste Groups Highlighted, Vishnumati 2001 46 Table 1.1 Number of Case Studies by Caste and Family Type 58 Table 2.1 Caste Hierarchy of the Muluki Ain 72 Table 4.1 Antenatal Care 118 Table 4.2 Place of Delivery 119 Table 4.3 Delivery Assistance 120 Table 4.4 Birth Preparedness: Women and Men 132 Figure 5.1 Trends in Modern Contraceptive Use among Currently Married 144 Women, Nepal 1996-2006

PRELUDE

A CONVERGENCE OF PERSONAS IN THE HINDU KINGDOM

In the early morning haze, on the day of Saraswati puja, a Nepali friend and I

rushed to catch a local “microbus” for the trip to Swyambunath stupa in Kathmandu to

write our names on the walls surrounding the Saraswati shrine. What could be more

auspicious for a doctoral student conducting research in Nepal than adding her name to

the countless names of students hoping to be blessed by the goddess of learning and

knowledge? That year, in 2005, I noticed that more names surrounding the shrine were

written in Roman script (instead of the Nepali Devanagri script) compared to what I had

informally observed the previous year. If only measuring social change was as simple as

comparing the percentage of names written in Roman script every year for Saraswati

puja. The ride to Swyambunath, however, was an even more poignant example of the

difficulty involved in making sense of what is “Nepali.”

At the bus stop just down the hill (pahaad - what would be considered a mountain

by people who had not grown up in the land of the Himalayas) we piled into a microbus

that was taking passengers to Swyambu. Our driver was waiting until the next micro for that route pulled up behind him, honking and insisting that our driver leave any other potential passengers for his profit. The micro appeared to be full already, and it happened to be full of Tibetan Buddhist monks. There were two monasteries in the community up the hill where I lived with my Nepali hosts and conducted my research,

1 2

and these young men had either just been visiting one of them or were in residence there.

They were fully robed in crimson and saffron dress. I was always impressed by the

matching accessories that monks found in those two distinct colors: that day it was

knitted hats and scarves for the morning chill. As a rule, it is imperative that as many bodies fit into the vehicles of public transportation as possible. And just when you think that number has been reached, two more people climb in. That day it was my friend and I

that crammed ourselves in amongst all the crimson robes.

As we took off down the hill south towards Kathmandu and then west towards

Swyambu, the driver inserted a cassette tape into the player in the front of what

Americans would call a minivan.1 I was expecting Hindi or Nepali pop music to begin,

but since our driver was particularly youthful anything from Guns ‘n’ Roses to Bryan

Adams would not have been unusual. I underestimated both the hipness of this young

man and the apparent increasing speed of global flows. Suddenly Eminem’s hit song

from the soundtrack 8 Mile (2002) began blasting. Thus I found myself in a microbus,

squeezed between monks, on the way to celebrate a Hindu religious holiday, listening to

Eminem. In amusement, I looked around half expecting to see one of the monks quietly

singing along.

I was the only one who felt disoriented by what, to me, was the discord between

the audio and other sensory information I was receiving: the contrast between the

Eminem song and the robed Tibetan monks surrounding me as we careened down a hill

in Nepal. This particular occurrence was distinct from the Nepalese appropriation of

1 “Microbus,” or for short, “micro” are words used in Nepali to refer to the small vans increasingly used for public transportation. They are faster than large buses, which have to stop more frequently to satisfy a greater number of passengers. 3

Guns n Roses or Bryan Adams, or the familiar “Jack and Rose” bandanas with the lead

characters from the movie “Titanic.” These popular culture items were all so commonly

encountered in Nepal, and the timing of their appearance so characteristically delayed,

that their very popularity seemed Nepali. It was the speed at which something popular in

the U.S. started to become popular in Nepal – and not just in the trendiest parts of

Kathmandu, in a surrounding rural area – that made this experience stand out. Yet to the others in the microbus, this scene held no irony. This is the kind of complex interrelatedness of which anthropologists must make sense in the age of globalization.

As Marcus and Fischer wrote, “Our consciousness has become more global and historical: to invoke another culture is to locate it in a time and space contemporaneous

with our own, and thus to see it as a part of our world, rather than a mirror or alternative

to ourselves, arising from a totally alien origin” (1986:134).

This dissertation is about women’s positions in the family and society in the

Kathmandu Valley: the experiences Nepali Hindu-caste women share, and all the ways

their lives differ. Inevitably it is also about globalization and social change, and how

women as individuals may come to embody that change through the processes of

reproduction. The scene in the microbus highlights the need to reveal the hegemony in

the flow of products, people, and ideas, and how people in a community utilize or reject

them. No new product, style, or technology warrants outward expression of surprise by

urban Nepalis. Some new miniature technological gismo can be seen almost daily in the

hand of a young Japanese model featured prominently in a photo in the mainstream

Nepali newspaper Kantipur. But which songs and products—or more relevant to the 4 questions here, what medical technologies and discourses on reproductive health—are imported and incorporated into daily life is related to larger issues of political economy.

Not only is the constant influx of material goods too common to be noticed, so is the juxtaposition of technologies in one locale such as a new motorcycle being stored in a ground floor room of a mud house that once housed the family’s cow, or a buffalo being butchered on the same rectangular slab of cement that serves as a community ping pong table built for local youth. Such seeming incongruities make it clear that contemporary anthropologists are always studying a mixture of cultural change and continuity, rather than just culture.

In the case of this research, my focus on women’s positions and health requires consideration of the inequities of existing cultural values and norms, and in addition, the unique configuration of the intersection of new medical technologies and family planning discourses with those norms. The following birth story, told by a middle-aged Nepali woman, exemplifies both of these aspects. The woman started going into labor during the time of year when Hindu Nepali women traditionally read a holy book called the

Swastaani. After giving birth, the woman and her entire household would have been in a state of ritual pollution (sutkeri), and so her husband was rushing to read more of the holy book before she gave birth; the state of pollution afterwards would preclude any further reading. She described the situation by saying,

Yes, it is extremely difficult [for women]. When one has that much of a stomach [meaning a large pregnant belly], how long will it take to die?! And also my husband is such a half-wit [literally half-done]. During the reading of Swasthaani, my husband called the neighbors to read aloud together from the Swasthaani thinking, "Oh, she's about to become sutkeri; there will be no reading of Swasthaani from tomorrow." And I was about to die downstairs. And from upstairs, [came the sound of] them reading aloud together the Swasthaani. That is so half-assed [ajkalTo]! They should have been worried, no? But there was no 5

[worry]. After that happened, I discovered how hard it was. When a person who should understand that it had become arduous [for me] does not, it is extremely difficult.

At first glance, this story appears to have very little to do with globalization. One could

easily explain the situation in terms of the patriarchal power structures and the

devaluation of women’s health, or the value of a Hindu tradition of reading a religious

text. These things are indeed at play. The project at hand, however, is to question such simplistic readings and to uncover how this story has everything to do with processes of globalization that influence reproductive options and the constructions of social hierarchy

such as caste, class, and gender. I portray a picture in which social change is intertwined

with demographic change.

Demographic change, in this case fertility decline, can be traced to the spread of

family planning messages, monies, and technologies combined with economic

development and urbanization. The introduction of internationally funded family

planning organizations to Nepal along with the embrace of development ideals about

lowered fertility by Nepali policy-makers beginning in the late 1950s and 1960s

eventually culminated in an institutionalized and state-sanctioned discourse about the

value of small families. It also resulted in the medical means for limiting and spacing

births: the provision of health services and contraceptive technologies. More broadly,

contact with products, media, and ideals from all over the world, paired with an increase

in exposure to new ideas through an increase in the level of education, has conditioned

young Nepalis to expect a world of competing interests and ideas. The extent to which they desire and are able to embrace new ideals given various structural constraints will be explored in terms of a fundamental activity central to family and social life: reproduction. 6

The findings presented here must also be taken in the context of specific historical events. The backdrop to this research was the Maoist insurgency and political unrest, a radically transformed social milieu from the past (romanticized) portrayal of Nepali life as peaceful and idyllic. Around the country the frequent violence caused by the insurgency in rural areas resulted in a noticeable amount of rural to urban migration. The violence may have caused an influx of families, particularly nuclear families, into the semi-urban study area – but this is difficult to determine because out of fear most people will not volunteer such motivations. However, I overheard people mention families moving to the area for this reason. In addition to the increased global flows and the potential for ideational change described above, a dramatic decrease in fertility was another structural component that was entwined with women’s experiences of procreation. I suspect that the Maoist insurgency had a significant effect on fertility decline in the past decade through urbanization and the separation of young men from their families, but conclusive research results are not available at the time of this writing.

After the unimaginable happened on June 1st, 2001 – the killing of the beloved

royal family – the political situation had slowly deteriorated until the point that the

international community began worrying and whispering about Nepal becoming a failed

state in 2005. His majesty the late King Birendra had been a beloved figure by the common people, and many believe his leadership and popularity sustained Nepal through its difficult years as a fledgling democratic constitutional monarchy full of corrupt political deals and characters. Despite its experiment with democratic principles since

King Birendra listened to his countrymen’s requests and granted a multi-party system of governance in 1990, Nepal was technically the world’s only Hindu kingdom because of 7

the continued role of the monarchy in governance and its control over the country’s army

until recently. After King Gyanendra, the much less popular brother of Birendra, took

over the monarchy,2 a triangular stalemate developed between the Maoists, the King, and

the political parties. The Maoist insurgency had already been going on for six years at

the time of the royal tragedy, but the occasional ceasefire and repeated peace talks

notwithstanding it began to intensify and gain momentum from that point through 2007.3

During my period of eleven consecutive months of field research beginning in

September 2003, there were frequent interruptions to daily life. Protest marches in the streets (julus), which sometimes involved huge crowds, created snarls of angry traffic as vehicles took to the back streets and side alleys in search of a way around the problem.

These protests often were peaceful demonstrations with signs and slogans, but sometimes involved throwing stones at security forces and police, sometimes carrying burning torches through the streets, occasionally ransacking university administrative offices, and almost always stopping traffic and making passersby nervous. The police met more aggressive demonstrations with tear gas and beatings with batons.

There were also frequent transportation and business strikes (bandh), sometimes called by the Maoists and sometimes called by the political parties. Outwardly a one-day bandh could seem like a holiday, with entire families spending the day together at home, shops closed, and children playing in the unusually quiet, empty streets. But the absence

2 Although conspiracy theories abound, the crowned Prince Dipendra was blamed for the shooting spree in the palace that killed much of the royal family. The prince then shot himself, and died shortly after being named King while in a coma. According to the rules of succession, Birendra’s brother, Gyanendra, was thus next in line for the crown.

3 In April of 2007, the Maoist rebels rejoined the government, moving back into mainstream politics. In September the Maoists left the interim government in order to apply pressure for the abolishment of the monarchy, and two months later rejoined as a result of Parliament approving the elimination of the monarchy. 8

of noise during the day always created a distinct and palpable eeriness. Anyone

operating a vehicle (other than emergency vehicles) did so in spite of the threat that they

would be attacked, or located and harassed at a later date. Much more than the one-day

bandh, several three-day and even a five-day bandh had dire implications for the

Kathmandu Valley. The great majority of produce that supports the million people living

there is brought into the valley by transport truck primarily along one major highway, and

a bandh lasting several days threatened food supplies. During those times the Newari family of which I was a part, since we lived in a more rural area above the valley on the northern rim, turned to the limited selection of produce in their backyard garden plot.

But those living in more urban areas had no place to turn. And as in all situations of

scarcity, the most impoverished people were the ones who suffered the most during the

bandhs, like those who depended on daily wage labor in order to purchase meals. The

Maoists also occasionally employed small bombs in various public places such as the

main bus station in central Kathmandu, in front of the Parliament building, and in forms

of public transportation in order to enforce a sense of fear. This description applies to the

Kathmandu Valley; what was happening in the rural areas, and especially the western

regions, was difficult to ascertain. Like dramatic changes in technology, these disturbing

political events and disruptions of life also became somewhat normalized over time, but

not without people regularly expressing their longing for the old days.

When I arrived in Kathmandu for two months of follow-up research on the

evening of March 31, 2005, the very next morning King Gyanendra issued a surprise

statement on the state-sponsored television channel declaring an indefinite state of

emergency which suspended fundamental rights like free speech, free press, and the right 9 to privacy. The international airport was immediately shut down, and all communication lines were cut, including telephone lines, mobile phones, and all internet connections.

Satellite phones were the only way to communicate with the outside world from within

Nepal. The King also sacked the coalition government and dissolved the Parliament, claiming that the political parties had failed Nepal by not resolving the matter of elections or the conflict with the Maoists. This was actually the second time Gyanendra had dismissed Parliament – a repeat of events from October 2002, down to the detail of dismissing the same prime minister. But this time because of the state of emergency and revocation of rights, the public understood the potential danger for even the common citizen. Human rights abuses abounded on the part of the king’s security forces and army as well as by the Maoists, and many people feared being kidnapped or “disappeared.”

Political figures, activists, journalists, and even professors were arrested and held without being told why or how long they would be kept in captivity. Family members were taken in the middle of the night by plain-clothed men. Reports of torture and rape were common. An Amnesty International press release dated February 10, 2006, stated that,

“Ten years of war and political instability have turned the human rights situation in Nepal into one of the worst in the world.”

In the community where I lived and worked, these events seemed strangely removed, although there were traces all around. The prime minister’s house, for example, was in sight from the rooftop patio of the house where I lived, and security was often tight on the main road leading past his house up the side of a mountain. The distance that separated the small town, literally at the end of the road, from Kathmandu below also sometimes seemed much greater than it was. One spring night when the 10

electricity had gone out (as it often did), I went up to the rooftop to take down my clothes

from the line on the flat rooftop patio (kousi) where they had dried earlier in the sun. As I

reached the top of the compact iron stairs, I gasped at the expanse of darkness. The entire

Kathmandu Valley was without power. I called out to my family, and my 23-year-old

sister said she had never seen that happen in her lifetime. Usually the lights from

Kathmandu twinkled like stars through the haze and smog in the valley below, and blackouts were confined to particular areas. I cannot describe the feeling of nervous anticipation and fear that that darkness engendered – it was if the entire city of

Kathmandu and its commanding sprawl were gone. We immediately feared that the

Maoists were somehow responsible, but we were wrong. I never did find out what

caused such an extensive blackout, but it was probably something much simpler and

more mundane than the possibility that sprung to our minds during such a time of stress.

So rather than the events seeming distant, it would be more accurate to say that

the fear of something bad happening was tucked neatly out of sight as people continued

on with their daily concerns and responsibilities. Most of the changes brought about by the insurgency and the emergency came in the form of inconveniences for common people in the community, and a feeling of foreboding, rather than violence.4 The armed

Security Forces in their blue camouflage uniforms along the streets of Kathmandu

became a familiar sight, as did the routine of checkpoints, strikes, and protests. Cadres of young army men riding in the back of a truck along the streets of the capital city would

position themselves facing outwards with guns poised, either ready for an attack, or to

4 This was not the case, as a stated previously, for people of high profile or people living in rural areas, and one should not make generalizations about the levels of violence based on my somewhat quiet, sheltered research area. 11

make an impression. This meant that for people on motorcycles, which outnumbered

cars on the road, there were rifles pointed directly at face level as they drove behind the

truck or passed it. It is unnerving to have a rifle pointed directly at one’s face from only a

few feet away as one rides on a motorcycle through busy, unpredictable traffic. On one

afternoon a man in street clothes with a bandana covering his face, armed with a rifle,

searched our house after claiming he and his partner were police officers on a routine check. These were the types of minor unsettling incidences that I experienced while

conducting research in the valley during troubled times.

This brings us back around to the scene on the microbus: my presence probably

deserves more interrogation than the Eminem song that I highlighted. To momentarily

shift the perspective away from my own, what was a koire (Nepali derogative term for

white people) doing on that microbus? And why was she joining the Nepali crowds at

the Saraswati shrine? My presence in Nepal as an American researcher, like the other

events and relations that I analyze here, was a result of my position in a larger system of

relationships infused with power. Despite my efforts as an anthropologist to fit in,

including dressing in kurta sarwaal and speaking Nepali, I would never gain membership

status. I would always remain the polite, but obvious, guest. My awareness of my

imposter status prevented me from actually writing my name beside the others for the

goddess. I pray that this will not, however, impede my success at honestly and accurately

portraying the lives of women in my research and creating an effective forum for their

voices.

INTRODUCTION

WOMEN, AGENCY, AND REPRODUCTION

Introductions

While conducting an interview in the afternoon sun on a rooftop patio of a shared apartment building, a neighbor interrupted to tell her unusual story. She boasted that many years ago she had gone to get sterilized after having only two children. Another old woman sitting with us expressed her surprise. “It’s true, I did,” the neighbor continued, “...and that day after I got the operation, I was lying there saying ‘oooh’ (from the pain), and at the same time my mother-in-law was saying ‘Maresh’ (may you die).”

She mimicked the chopping gesture of her mother-in-law for humorous effect, and the group of women on the porch laughed. Her father-in-law had called her sinful because of her actions, and even her doctor had not approved at first. When she arrived to get the operation, the doctor asked about her children. When he found out she had only two children under five years of age, and only one son, he advised her not to have the operation because there was no guarantee that the children would survive. After he refused, she pleaded, “…if they die, let them die; it doesn’t matter to me… Do it, do it for me.” She claimed that through her own perseverance and boldness, she convinced the doctor to do the operation.

This story reveals several things about the context in which the women in this study live: the reality of infant and child death, the practical and moral value of children

and especially sons, and the increasing level of medicalization of women’s reproductive

processes. The example is unusual in the boldness with which the woman acted given her historical and cultural location. She was telling the story of her actions of around

fifteen years earlier – a time when the dominant discourse held that women should have

five or six children and two sons. Her actions at that time were an uncharacteristic

example of women’s agency and going against the wishes of others, and that is precisely

the reason she interrupted us to tell her story. Hers was an exceptional case. She was

proud of and amused by her act of defiance.

At the broadest level, this research is about the contradictions with which people’s lives are filled. I argue that the contradictions – whether conflicting desires expressed by individuals, disagreements between generations or genders, or differences between things said and things done – are a productive area for the investigation of social change.

Rather that trying to smooth over contradictory data in social science research, I propose

that a thorough exploration of the contradictions will result in a more theoretically

sophisticated understanding of social change. This approach is similar to the one used by

Harlan and Courtright, focusing on areas “where the idealized constructions reveal their

flaws and vulnerabilities as well as their enduring capacities” (1995:4); but my approach

differs by studying contradictions found in mainstream people and culture rather than

those “at the margins” or in the exceptional opening story. I embrace the contradictions

found within individual case studies of a range of average, everyday people with the goal

of extracting from these examples their relevance to social change. Change, after all,

results from questioning, contradicting, altering the status quo, even in small increments.

What remains is how to capture and analyze the dynamic process of change in culture

and society.

In the pages that follow, I tackle the challenge of analyzing culture and inequality

and the embodied experience of social hierarchy among Hindu-caste Nepali women. I

address, layer by layer, the structural components of Nepali society that motivate and constrain women’s agency. I begin by exploring the ways that caste and class intersect with gender norms and divide women’s experiences. I propose that poverty has the power to diminish these moral systems of hierarchy in specific ways. Afterwards, I further complicate an understanding of women’s experience of gender by taking a life course perspective and demonstrating with illustrations at the micro level the significant changes in status that occur over a lifetime in culturally determined ways. After accounting for and revealing the complexity and interaction of these layers of social hierarchy, I turn to the experience of procreation. Using women and reproduction as the subject matter, I provide a window for how the dialectic of agency and structure plays out in individual lives. Moreover, I argue that women actually embody social change through reproductive processes: in a phenomenological sense with the bodily experiences of pregnancy, miscarriage, birth, and post-partum recovery; and in an ideological sense in terms of their struggle with cultural expectations placed upon them and their bodies as reproducers.

With the aforementioned social hierarchies as a backdrop, the thesis culminates in a description of women’s specific historical, local yet globalized, position in terms of a major influence in the experience of reproduction: their location in the midst of dramatic fertility decline and the accompanying struggles of social change. Fertility has declined

rapidly in Nepal from 4.6 births per woman reported in 1996 to 3.1 births per woman reported in 2006 – a drop of one and half births per woman in less than one generation.

Myriad factors may have contributed to this notable decline, including those predicted by theories of fertility transition such as urbanization, modernization, and the increased availability of contraception. But among the salient factors are also ones related to the specific historical circumstances of Nepal in the last decade. The ten years of political insurgency by the Maowadi that affected these fertility rates (1995-2006) caused internal and external displacement of people through the migration of families from rural to semi- urban and urban areas and the separation of men from their families through internal and international migration. In semi-urban areas of Nepal like the one in this study, the ideational change regarding the ideal number of children that must accompany fertility decline is inadequately understood.

The women described and quoted in the following chapters were part of 27 case

studies in a research project carried out over a period of thirteen months during 2003-

2005 in a semi-urban village in the outskirts of the Kathmandu Valley. They are Hindu-

caste Nepali women who were selected from a household survey of two wards for their representation of a variety of caste levels, socioeconomic statuses, education levels, and

household compositions. They all shared the statuses of wives and mothers. What is

gained, and what is lost, through focusing on women, their households, and their community?

Defining Women

Women in developing countries have been defined in various ways over the past

few decades according to the overriding theoretical assumptions and dominant players of

the time. While the deconstructionist emphasis on individuality and on the aspects of a

person’s position that cross-cut a supposed group such as “women” seems obvious by

those of us situated in the early 21st century, an archaeology into the past of the ways

Third World women were constructed and defined by demographers and development

planners is useful in understanding the globalization of historically contingent, western ideas about population growth and the technologies that followed those ideas. I do not intend to pass judgment on demographers and development agencies, but merely to trace the ways in which a concern with human population growth developed and became reconstituted in terms of a concern for reproductive health and women’s empowerment. I

highlight the similarities between this progression and the move within feminist theories

from an initial western, white solipsistic stance toward the shared qualities of “women”

around the world to fractionated visions about what is shared and what is unique among

women of different ages, ethnicities, classes, sexual preferences, etcetera. And

furthermore, I draw connections to the historical transition in anthropology from the

tendency to identify people in other cultures as objects of study in the Enlightenment

fashion to problematizing the authoritative ethnographic stance and even the possibility

of objectivity. Major developments in postmodern philosophy underlie and thus connect

these three strands.

The three strands reflect both an increased sensitivity to difference and a growing

suspicion of the privileged status of the objectifying language of science. Postmodern

philosophy throws into question the possibility that any language – scientific or otherwise

– could be compared to non-linguistically constituted “facts” about the world. If such a comparison is shown to be impossible, then the objectifying language of science is opened to historical deconstruction (see Derrida 1976) and charges such as it is an epistemological deployment within social power relations (see Foucault 1966). In terms of family planning initiatives, this is reflected by a radical shift in a self-interested concern with overpopulation and how developing countries were threatening global (read western/industrialized) interests to a focus on women’s health and status. The recognition of the symbolic and contingent nature of language enabled the deconstruction

of categories like “women” or “the other.” In this section, I address each of these strands

briefly and how my approach to studying women and reproduction is necessarily situated

within them.

Population Problematics

In the realm of population and development planners, women were first equated

with fertility itself: as something that needed to be controlled. During the 60s and 70s,

reviving Malthusian logic and explaining scarce resources and poverty in terms of an

unchecked human population “explosion” (Ehrlich 1971), some activists fueled popular

support of programs aimed at reducing fertility in developing countries. It did not take

long for opponents to criticize this limited perspective by proposing that

overconsumption in many so-called developed nations and an unequal distribution of the

world’s resources were what was driving the hunger and suffering evident in the world’s

poorest nations (Hartmann 1995). The early history of the family planning movement

was blighted by examples of forced sterilization, Norplant testing, and an assumption that

user-controlled methods like the birth control pill would not be as effective in developing

countries. Demographers tried to distance themselves from the activists on both sides of

the issue (see Mason 1995) and the political agenda of limiting births in developing

countries, yet the source of much of their funding came from governments and

international agencies that had similar agendas (Greenhalgh 1996).

In the 1980s and 90s, women in developing countries began to be portrayed by

social science researchers as more than fertile entities that needed controlling; women

were rational, thinking, feeling beings who were influenced by testable environmental

factors. This represents a significant break from viewing women synonymously with

fertility. Women were no less the objects of “biopower,” in Foucauldian terms (Foucault

1978), of both the state and international population and development agencies, but they

gained an increase in the recognition of their human dignity. For example, sterilization

campaigns and experimental trials in developing countries were no longer fathomable,

but the control of populations and fertility remained a valued goal: the hegemonic

aspiration of lowering fertility continued to drive much population policy and funding.1

Scholars turned their focus to defining the concepts of women’s status and autonomy, and testing the relationship of variables such as education, control over resources, and employment outside of the home in a search for the salient variables that lead to higher status.

Concurrently in the 80s and 90s, there began an attempt, primarily on the part of anthropologists and sociologists, to incorporate theories of agency and new theoretical

1 In line with Judith Justice’s critique of bureaucratic institutions in Policies, Plans, and People (1986), I do not intend to criticize the intentions or motivations of individual people but rather draw attention to the collective goals as stated in program outlines (for my overview of such policies see Chapter 5).

19 developments in feminism2. Women were viewed as actors, as individuals imbued with the ability to make decisions and take action. This was followed by a somewhat inconsistent and muddled usage of the word “agency” in the social sciences to mean anything from free will to autonomy. Studies and stories of resistance by oppressed groups of all kinds became popular, and a much needed corrective to a history told by the dominant took the form of subaltern studies. Attempts to apply the concept of agency cross-culturally spurred an exploration of the assumptions surrounding “choice” and universal human rights, particularly in regard to the oppression of women and minority groups.

The deconstruction of the monolithic category “women” was a triumph within feminism by women of color, and it served the development of Third World Feminism

(Mohanty et al 1991; Narayan 1997). Such a postmodern approach, the study of difference, and the promotion of a whole host of women’s voices, was echoed in anthropology in the interpretive approach championed by Geertz (1973) and the innovations in representation and multivocality called for by Clifford (1986). In response to such western, individualist notions of cultural relativism, a growing concern with universal human rights is at the center of current discussions of transnational feminism and of influential philosophical works like Nussbaum’s Women and Human Development

(2000). Along with my analysis of women’s well-being, it is to this debate that I hope to contribute with the present project.

2 See Nancy Riley’s critique of demography in this regard (Riley 1999).

Status and Autonomy

The concept of “women’s status” in developing countries became a topic of

interest among western scholars primarily because of its connection to fertility decline.

The social and institutional denial of women’s equal access to and control over resources

were associated with high rates of fertility worldwide. Accordingly, researchers

documented the significance of discrepancies between men and women in relative

autonomy, education, resources, and occupation for the achievement of general well-

being around the world. The improved status of women was hypothesized to result from increased educational or economic opportunities (Caldwell 1982), both leading to fertility decline. In an effort to specify these associations, homogenous and unidirectional models to explain improvement in the status of women were challenged. Mason (1986, 1987) was the forerunner in pointing out the failure of social demographers to recognize the multidimensionality of women's status and its variation across social "locations," for example differences in status in the household versus in the community, or across the various stages of the life cycle.

The growing concern for the status of women across the globe was evidenced within the development world by the 1994 International Conference on Population and

Development (ICPD) by the “empowerment of women” becoming the catchphrase of the conference and by the presence of a large contingency of organizations involved in women’s rights. The resulting ICPD Program of Action recognized reproductive health and rights, and women’s empowerment more broadly, as cornerstones of a successful population and development program. The following year in Beijing, the Fourth World

Conference on Women, as its name implies, focused more broadly on women’s rights.

The platforms established at Cairo and Beijing “converge in their affirmation of women's human rights and the recognition that solving the world's most pressing problems demands the full participation and empowerment of the world's women” (United Nations

Population Fund).

After Cairo and Beijing, scholars continued to refine the ways they talked about women’s status. Change in women’s status may be multidirectional, improving in some realms and worsening in others, as traditional norms and behaviors that protected women break down (Bradley 1995). In regard to relative life expectancies between men and women, Amartya Sen pointed out (1990) that economic development is often accompanied by a relative worsening in the survival rates of women. In India women’s relative positions deteriorated with the onset of economic development for decades until the gap between the life expectancy of men and women narrowed within the last decade or two (Sen 1990). In such cases, women may not have access to or benefit from advances in health services and infrastructure and general economic progress. New practices can also reinforce established attitudes towards gender roles. Laura Ahearn’s ethnographic research in a Magar village in Nepal illustrates how increased female literacy rates, for example, do not automatically lead to better lives for women (Ahearn

2001). While education may be related to a decrease in fertility, it is not necessarily related to an increase in power for women (Riley 1999). Education has different meanings and uses in various societies; thus education and status will not have identical relationships across cultures.

Life course stages are particularly salient for the position of women in Hindu cultures in South Asia. A woman moves from being considered the embodiment of the

goddess as a young daughter, to a sexual being that is simultaneously a threat to her husband’s patriline and the source of its continuation (Bennet 1982), to an asexual post- menopausal being with few gender restrictions (Das 1992; Lamb 2000). Within the

reproductive years, women’s positions also vary dramatically within the household family system. In much of Nepal and northern India the predominant pattern of marital residence is patrilocal with multi-generational extended family living arrangements

(Fruzzetti 1982; Fruzzetti et al. 1992; Vatuk 1975). The new wife bears the greatest burden of domestic duties and has the lowest status in the hierarchy of the family. Over time, her position improves with the birth of children, especially sons, and the introduction of another young wife into the family by her husband’s brother or eventually by her own son (Das Gupta 1995). Das Gupta (1995) has demonstrated that in northern

India, an area somewhat culturally similar to middle and southern Nepal, fertility preferences and maternal and child health differ dramatically according to the life course status of the woman and the sex of her children, when controlling for established predictors such as education, wealth, and caste.

Like women’s status across life stages and social locations, households are also dynamic in several ways. Given the importance of the influence of family and household on the position of women over the life course, this project addresses the variation in women’s positions caused by the different patterns of household composition that are becoming more prevalent in Nepal. Change in household composition over time has important demographic implications (Kertzer 1991; Fricke 1994), and households have diverse, historically specific patterns (Yanagisako 1979). Historical changes also affect individuals differently depending on their life stage, and earlier experiences affect later

events in the individual's life (Kerzter and Hogan 1989). In his extensive research in

Nepal, Axinn (Axinn 1992; Axinn & Yabiku 2001) documented a shift from a familial

mode of organization of life’s fundamental activities to a non-familial organization of

activities. Using the familial mode of organization framework to connect macro-level

social change and individual-level childbearing behavior, he concluded that both early

childhood and contemporary adult community context affect behavior. "This conclusion

points toward contextual models of behavior that incorporate a much more

comprehensive view of contextual change over the life course than most social science models have thus far considered" (Axinn & Yabiku 2001:1252). Building on this, I focus on how the culturally determined trajectories of women’s married lives vary according to household composition by examining the webs of relationships in which women’s lives are situated. Variation in household structure affects the motivations,

opportunities, and constraints on women, especially through the presence or absence of kin and the dynamics of relationships within the household.

As more and more research was conducted, it became increasingly apparent that women’s status was a complex thing to capture or measure. Despite progress in the way social scientists were defining women’s status, it remained a somewhat limited concept because it was most often a measure of individual, or occasionally structural, inequality and was not well-suited to account for women’s agency or the contingent and fluid nature of social processes.3 Depending on their theoretical stance, social scientists proceeded

either by aiming to operationalize the concept more effectively or, alternatively, to dispense with any such universal category. I turn now from the project of tracing the

3 With several anthropological studies being notable exceptions, such as Ahearn’s (2001) and Fruzzetti’s (Fruzzetti and Tenhunen 2006) in South Asia

definition of women from the perspective of social scientists concerned with development

or fertility decline, to quite a dissimilar endeavor – the deconstruction of the category

“women” by feminist theorists.

Diversity in Women’s Lives and Voices

As scholars subscribing to a positivist theoretical approach were refining

categories such as women’s status in the 80s and 90s, feminist theorists were busy

deconstructing such universal categories as “gender” and “women.” Feminist agendas have moved from universal questions, categories, and arguments toward considering

particular contexts (Errington 1990). Within anthropology, this translated into moving

away from examining the worldwide reach of universal subordination theory, the

man/culture woman/nature analogy, and the separation of domestic and public spheres

(Ortner 1974; Rosaldo 1974). Feminist anthropologists began focusing on ways the

definitions of “woman” and of “gender” change historically and across cultures.

Differences within a single society were also brought to the forefront. The

protests of women of different classes, ethnicities, and sexual preferences had made

Euro-American feminists aware of the multitude of almost entirely separate realities

experienced by different women. As di Leonardo wrote, feminist scholars were forced to

confront “…the question of ‘difference’ – the multiple racial, ethnic, class, sexual, age,

regional, and national identities of women – as they noted their own restricted

demographic representation and research interests” (1991:18). Prior to that point, women

had been discussed as if they were a unified assemblage – as if being female alone was

uniting enough to bind them together in a definitive group.

The two example questions of the new focus on particularities and difference that

Errington posed are representative of the scripts guiding much recent research on women

and gender, including this project. First, “How do race, class, and gender intersect to

affect the welfare of women in such-and-such society and historical period?” And

second, “What is the dominant prestige system of such-and-such society, and how does gender difference fit into it?” (Errington 1990:9). Ortner and Whitehead (1981) wrote

that societies’ “prestige structures” will shape the constructions of gender in each society,

and I demonstrate how that plays out in Nepal.

What do these directions in feminist theories mean for the study of women’s

status? As early as Quinn’s 1977 review of the status of anthropological studies on

women’s status, anthropologists suggested treating “women’s status” as several different variables, implying that the category was too broad. Quinn wrote that “such

interdependencies” among different aspects of women’s status “must be specified, not

assumed” (183). Her assessment has continued relevance for research today. Rather than solely depending on measures of variables like education and influence in household decisions to assess women’s status, I examine the “intersectionality” of key aspects of women’s lives because variables can interact in surprising ways in certain cultural contexts. Ranking high in one prestige system and low in another prestige system can shape or challenge constructions of gender in interesting ways. In this research, economic hardship acts as an equalizing mechanism for women of different castes, in some ways releasing women from the gender restrictions of their high caste, but at the expense of conditions that come with economic security such as a more comfortable life with fewer injuries, illnesses, and malnutrition. I illustrate the tradeoffs in different

situations for women in which certain aspects of their well-being may be forfeited or

suppressed while others are gained or improved. It is difficult to place a value on these

different aspects, and therefore hard to compare them.

In her comprehensive review article on what feminist contributions could offer to the field of demography, Riley concluded that evidence and research called for

“culturally grounded research on gender that does not assume that western notions of any institutions or concepts, including gender, status, and power, are necessarily similarly

valued or interpreted in societies” (1999:382-383). As Rosaldo stated, “women’s place in

human social life is not in any direct sense a product of the things she does, but of the

meaning her activities acquire through concrete social action” (1980:400 emphasis

added).

Following such logic about cultural specificity, Errington argued (1990) that the

categories “power” and “status” cannot be applied cross-culturally. While the notion of

universal categories of power and status are not entirely satisfactory, and while there are well-documented exceptions to the primarily economic and control-based definitions of these two categories (Atkinson & Errington 1990), I do not think that these categories should be rejected outright. Abandoning such terms or categories on the philosophical basis of the cultural constructionist argument would mean that I am left with a postmodern pile of unique cases. Similarities and differences among individuals or groups are partly a matter of scope, and I certainly do not believe that my case studies or the larger population in this study are unique. When compared to other groups in other studies, the similarities among my case studies become apparent. Errington similarly conceded that in Southeast Asia the differences between women and men were not highly

marked socially “if they are contrasted with differences in other parts of the world rather

than taken on their own terms” (1990:5). I like to think of broad categories like gender

and women’s status as starting points to investigating the meaning or even the existence of such concepts in other societies. As long as such categories are recognized as constructs and are not naturalized, as long as they are problematized whenever it is necessary to use them, I think they can be useful in maintaining a sense of perspective among various cultural contexts.

Thus I find myself unable to escape the comparative project (for without it how does one maintain a sense of perspective – or, more accurately, be aware of multiple potential perspectives?), but equally unwilling to claim total objectivity. Instead, I aim to produce what Haraway called (1988:582) “situated knowledge,” which is grounded in where I am positioned as a researcher. Haraway’s influential essay “Situated

Knowledges” (1988) was an “argument for situated and embodied knowledges and an argument against various forms of unlocatable, and so irresponsible, knowledge claims.

Irresponsible means unable to be called into account” (583). Feigning to be completely objective (when it is impossible) is thus deemed irresponsible.

Haraway was able to find some middle ground in the debate over science and objectivity in her persuasive essay. There is always a perspective involved in research, even if it be one resulting from training in the epistemology of western science. “There is no unmediated photograph or passive camera obscura in scientific accounts of bodies and machines; there are only highly specific visual possibilities, each with a wonderfully detailed, active, partial way of organizing worlds” (583). Accordingly, I am presenting an unavoidably partial account – a situated knowledge, but one that is grounded in over a

year’s worth of first-hand experience and interviews and steeped in the literature

produced about and by this particular region. I attempt to acknowledge what I can and

cannot see from my perspective – as an individual, and as a trained anthropologist. Far

from an exercise in narcissism, this reflexive approach endeavors to expose the socially

conditioned structures that underlie epistemological approaches to and the formulation of

theories about the social world (Bourdieu and Wacquant1992).

Reproduction, Agency, and Social Change

There is a danger inherent to examining female corporeality because of a history

of patriarchal thought that confined women to the biological role of reproduction, coding

them as being more natural and corporeal than men (Grosz 1994). In focusing on

women’s experiences of procreation and their expression of agency through procreative

acts, I risk essentializing women as child bearers or at the very least overemphasizing their role as reproducers. Historically this has been a defining aspect of womanhood in this cultural context, with the culturally acceptable exceptions of becoming a nun or an ascetic (Khandelwal et al 2006). My focus on reproduction also reflects the continued dominance of this role in mature women’s lives as they themselves tell it. Now, on the contrary, fertility decline in Nepal has provided young women with a greater than ever level of opportunity to define themselves as something other than/in addition to reproducers. It is precisely this historical and structural location that sets up a relevant and timely study of women’s embodiment of social change.

In the decade following Ginsburg and Rapp’s assertion of the value of studying reproduction to social theory (1995), anthropologists have supplied abundant

documentation to support it. Scholars have demonstrated that the politics of reproduction may serve as an entrée into the projects of defining self and society in a world of relentless technological innovation, competing discourses, and social change. Recent developments in social theory recognize the dynamic nature of social processes as resulting from the constant dialectic between individuals and social structure (Bourdieu

1977; Ortner 1984). The ideas behind statements like this one, that “people everywhere actively use their cultural logics and social relations to incorporate, revise, or resist the influence of seemingly distant political and economic forces” (Ginsburg and Rapp

1995:1), resonated so widely with anthropologists that they have become almost ubiquitous in contemporary academic writing. One aspect of this direction in anthropology serves as an underlying guiding question throughout this project: what reproduction, as a domain of inquiry, has to offer toward illuminating ways of studying process and social change. I argue that the reproductive bodies, beliefs, and behaviors of

Nepali Hindu-caste women tell us about the gradual acceptance or rejection of new ideals and social hierarchies.

In order to avoid the limitations of a static, universal category of women’s status noted in the previous review, I take into account not just women’s structural positions but also their active positioning, and consider how both change over the life course. In much of South Asia, women’s positions are situated in a web of familial and social relationships, and are best understood within this context (Fruzzetti et al 1976a, 1976b).

In regard to fertility and health, women do not act alone but rather as part of this web of relations – as part of a social group. This is not to deny women agency. Examples of women’s resistance to structural constraints and motivators are an integral part of the

social reproduction of cultural norms. A better term is needed that can encompass both

the relational/structural position of women and women’s agentive positioning, but for

lack of such a term “position” will be used as both a noun and a verb to imply the

dynamic web of relationships and strategic action.

The way I use the word “position” intimates my approach to defining the concept

of agency. I side with Ahearn in her rejection of defining agency either as free will or

resistance (2001). The former “ignores or only gives lip service to the social nature of

agency and the pervasive influence of culture on human intentions, beliefs, and actions”

(2001:114). Using agency as a synonym for free will ignores the part of the dialectic of

social being in which anthropologists historically have specialized: variably defined as

culture, social structure, structure, the system, and discourse. With the popularization of

the interpretive or symbolic anthropology of Geertz (1973) and Rabinow (1977), notions

of culture as fixed, bounded, stable and systematic structures became defunct. Post-

structuralist accounts swung the focus in anthropology away from systems and structure

toward the other direction of individuals and praxis. The practice theory that developed in the late 1970s and 1980s (Bourdieu 1977, 1990; Ortner 1984) emphasized a new focus on individuals and agency, but in a more balanced way, as part of an ongoing process of mutual influence and production between agents and structure. The study of practice was not developed as an alternative to studying systems or structure, but as a way of incorporating actors into the equation (Giddens 1979).

In sum, I share with the “newer practice theorists” as Ortner called them in 1984,

“a view that ‘the system’ does in fact have very powerful, even ‘determining,’ effect upon human action and the shape of events” (146). Likewise, as Ortner described, more

often than not I view the shaping power of culture or structure as a somewhat insidious

force, particularly because I study inequality, power relations, and social hierarchies. I

felt the presence of such power hierarchies more than ever before while living and doing

fieldwork in Nepal. Even casual, brief conversations with strangers involved a certain

jockeying for status, or at least a process of figuring out one’s position vis-à-vis the other

person. Different rules govern interactions and even forms of speech depending on who

is senior and who is junior, so usually it is merely a necessary exercise in order to

maintain appropriate politeness in most situations. But a certain amount of savvy is also expected of people in this cultural context, even in the simplest of business transactions such as bargaining for an apple on the street. And those who lack it are quickly discovered and dominated in the interaction. Thus even in informal, mostly anonymous interactions, a certain amount of sizing up of one another occurs.

When studying a group that is subjugated within the larger society, “culture” can start to seem quite oppressive, and even synonymous with hegemony. “To say that ‘men’

[sic] define and shape their whole lives is true only in abstraction. In any actual society

there are specific inequalities in means and therefore in capacity to realize this process”

(Williams 1977:108). In this light not only is culture a system of constraints upon

individual agency, it is also a system that obscures the domination of some groups to the

advantage of others. Bourdieu’s notion of habitus as a “deeply buried structure that

shapes people’s dispositions to act in such ways that they wind up accepting the

dominance of others, or of ‘the system,’ without being made to do so” (Ortner 2006:5)

acknowledges the centrality of social relations of power, but he does not address how

these play out in terms of individual actors on the ground. One issue worth considering,

32 for example, would be the fine line between constraint and inspiration. How ought one judge motivation, in negative or positive terms? Do treasured cultural values inspire women to produce sons, or does a patriarchal society demand that women produce sons and limit their roles in society to that of childbearer? Are actions that are in alignment with the patriarchal structures in place an example of choice, of agency?

In this project, I adhere to Ahearn’s fine tuning of the definition of agency.

“Agency,” in her succinct definition, “refers to the socioculturally mediated capacity to act” (Ahearn 2001:112). Some feminist scholars have argued that agency must be defined by action, but Ahearn locates agency in the capacity for action rather than the act itself. From this standpoint, limiting the definition of agency to resistance is theoretically deficient. And for some anthropologists, this broader meaning is part of the appeal of the concept of agency: “…it is free from the ideological and conceptual understanding of the notion of resistance, and therefore more adept at capturing the fluidity of transformative act” (Fruzzetti & Tenhunen 2006).

Toward a Critical Phenomenology of Procreation

Reproduction is a site of struggle by individuals, couples, families, and even nations and international organizations, to define values and goals. All of these players have interests at stake, such as giving or not giving birth, giving birth to a certain sex, and controlling morbidity, mortality, and fertility. Social, familial, and gender norms are inscribed on women’s bodies through the physical processes of procreation.

Concurrently, if women experience these norms bodily, then they are also capable of acting and exerting agency with and through their bodies. I aim to capture both sides of

this process of gradual social change through analyzing women’s conflicted feelings and

behavior related to reproducing.

The ongoing changes in social structure, the expansion of biomedical services,

and the promulgation of the family planning discourse in the Kathmandu Valley provide

a backdrop for the analysis of the ways Nepali women negotiate and reproduce the

hierarchies within which they define themselves. Nepalis share this position with women

throughout the developing world wherever the purported value of fewer children, along

with the means to limit reproduction, has been imported. Specific cultural and

demographic circumstances have combined to create a situation for young married Nepali

women where they have fewer chances to fulfill the expectation to have a son. I use the

debate over sons to move between a broad overview of recent change and continuity in

the context in which women find themselves, and a close-up of the change and continuity

that women themselves engender. A generation of women is questioning the necessity of

sons or multiple children; and women and their bodies are the locus of cultural change.

Addressing these topics requires a consideration of the body that transcends the

dualisms historically present in mainstream ways of thinking about the human body.

“The body is a most peculiar ‘thing,’ for it is never quite reducible to being merely a

thing; nor does it ever quite manage to rise above the status of thing… If bodies are

objects or things, they are like no others, for they are the centers of perspective, insight,

reflection, desire, agency” (Grosz 1994:xi).

Csordas went beyond anthropology of the body, which he described as merely studying the ways in which cultural values are “inscribed” on the passive body, to understand embodiment as the existential condition of cultural life. He emphasized the

distinction between the body as an empirical thing or analytic theme, and embodiment,

which he defined as the “existential ground of culture and self” (1994:6). From this

standpoint, embodiment is not reducible to representations of the body, to the body as an

objectification of power, to the body as physical entity or biological organism, or to the body as inalienable center of individual consciousness. The purpose is to contribute to a

phenomenological theory of culture and self – an anthropology that is not about the body

but is from the body (Csordas 1994).

I find this to be a useful starting point in the study of reproduction; a state of

exploring experience, perception, and knowledge prior to the divide of the objective and

subjective. Merleau-Ponty, the French philosopher and influential figure in outlining

phenomenology, suggested starting analysis prior to the division of the objective world

and experiencing subject in order to gain the “experience of perceiving in all its richness

and indeterminacy” (Csordas 1990:9). As Byron Good suggested in Medicine,

Rationality, and Experience, an objectivist rendering of the body does not work well

when it comes to a process such as birth, which is not able to be abstracted from its social

and symbolic aspects (1994: 117).

Yet the study of reproduction necessarily includes a consideration of the role of

hegemony in reproductive lives (Greenhalgh 1995, 1990; Ginsburg & Rapp 1995) or the

aspects of the body politic (Scheper-Hughes & Lock 1987) related to procreation, in this

case ranging from the personal politics of women’s relationships to the global politics of

the messages and services of international population programs. I engage only briefly

with global influences through the description of international family planning programs

and fertility trends and differentials at the national level, with the majority of evidence

coming from an in-depth study of individual women and their families. I place their

individual struggles in the context of larger structural factors, and take their experiences

as indicative but not statistically representative of other similarly positioned women.

The challenge in studying individuals and fertility behavior lies in “giving

demographic actors agency without endowing them with a utility-maximizing rationality”

(Greenhalgh 1995:19). Reproductive decision-making, much like human agency, “is correctly understood not as a sequence of discrete acts of choice and planning, but rather as the reflexive monitoring and rationalization of a continuous flow of conduct” (Carter

1995:61). Ultimately the “decision” to give birth in this research context is more a process than a moment, one that involves power dynamics between the sexes and among family members. Likewise, it is shaped by the structural constraints applicable to that context. The kind of tension between agency and structure writ large is manifested at the microlevel, or “up close,” in the state of internal conflict that Nepali Hindu-caste women expressed over fertility decisions in this research.

Feminist anthropologists were leaders in questioning the objectivity of science and the authority of anthropologists as interpreters of culture. As Haraway proposed in the late 80s, “We need to learn in our bodies…” (1988:582). “Feminist objectivity is about limited location and situated knowledge, not about transcendence and splitting of subject and object. It allows us to become answerable for what we learn how to see”

(1988:583). In an attempt to mitigate my position of authority in this text, as researcher, interpreter, and author, I provide quotations from interviews with various women captured on audio tape. In an effort to avoid a sense of timelessness, I often give contextual cues as to when the conversation was taking place, since people’s level of

comfort with (and fondness of?) me and my questions fluctuated throughout the research

period. I also acknowledge a performance aspect of the interviews, with this aspect being

more evident in some scenarios than others, and note when other people are present during a discussion.

In my analysis, I work to maintain a sense of open-endedness to people’s

statements and decisions in keeping with my focus on contradictions, process, and

change. Ultimately, I cannot truly know what these women were experiencing; as

Haraway wrote about subjugated perspectives, “There is a premium on establishing the

capacity to see from the peripheries and the depths. But here there also lies a serious

danger of romanticizing and/or appropriating the vision of the less powerful while

claiming to see from their positions” (1988:583-584). Although I do not make claim to a

“correct” interpretation of women’s comments and lived experiences, I do identify

constraints and possibilities for action. Ahearn emphasized that “…it is the task of the

linguistic or cultural anthropologist to identify the constraints on the range of

interpretations that might emerge from any discursive event,” thus shifting our focus

away from a search for definitive interpretations (2001:56).

Taping, transcribing, and translating hours upon hours of interviews is a time

consuming, expensive, and painstaking process – but because it is one of the few ways I

could allow women to speak for themselves in this text I pursued it. The endeavor to

emphasize multivocality in anthropology is an inadequate one as of yet, given that written

texts are still the prevailing form of relaying information about research in anthropology.

Videos and hypermedia have the potential to overcome the barriers to more extensive

multivocality, but the need to protect the identity of participants in studies often makes

such methods unfeasible. This being the case in my research, I rely on passages of

women’s own words translated into text in order to give voice to a variety of viewpoints.

And this is a fortunate thing, for often their stories are more poignant and their teasing

more entertaining than any summary that I could produce.

Women Experiencing and Producing Social Change

In summary, this research utilizes anthropological theory and ethnographic and

demographic methods in order to explore the ways in which household composition,

caste, and class interact and affect women’s positions, and subsequently their fertility and

health, in Nepal. From a practical point of view, healthy, capable children are important

for a woman’s immediate status within the family and her long-term security in much of

South Asia. In this context, the lived experiences of pregnancy, childbirth, child rearing,

and sometimes child loss, are defining aspects of womanhood. These processes are

formative in a woman’s identity and status, and her understanding and experience of

body and self, within the cultural context. Equally as important to understanding the

construction of womanhood are aspects of social structure such as gender roles, kinship

relations, economic stratification, and household structure. This research builds upon theoretical traditions that link these two levels, relating the lived experience of women with structural motivations and constraints.

In this introduction, I present an overview of my approach to issues of women’s status, social change, and health in developing countries through the theories and

methods of feminist studies, critical medical anthropology, phenomenology, and demographic anthropology.

Chapter 1 provides an overview of and rationale for the ethnographic and quantitative methods that I used in this research. It also introduces the research location and discusses ethnographic encounters. In Chapter 2, I describe how systems of social hierarchy in South Asia have left a legacy of inequality that relates directly to fertility and maternal health. I explore the ways that caste, gender, and economic hierarchies interact and conflict, producing some surprising results.

After looking at the ways social hierarchies affect women’s lives, Chapter 3 transitions to the subject of familial hierarchies and the life course. I exhibit the dramatic ways that a woman’s life changes as she moves through the stages of being a daughter, a wife/daughter-in-law, a mother, and a mother-in-law. I end with a consideration of family systems and the household cycle. Joint families undergo a constant cycle of splintering and reforming, “like a potter’s wheel” in the words of one informant. I review the reasons for forming a nuclear family, and the implications of living in a nuclear household for women and fertility.

In chapters 4 and 5, I argue that the burden of both caste and gender inequality affects women broadly in terms of the pressure to reproduce, and more acutely during the physical experiences of reproduction. Nepali women are embodying social change through reproductive processes: in a phenomenological sense with the bodily experiences of pregnancy, miscarriage, birth, and post-partum recovery; and in an ideological sense in terms of their struggle with cultural expectations placed upon them and their bodies as reproducers. Recent trends in Nepal toward fertility decline and increased opportunities for women have placed contemporary women of reproductive age in a bind between the

“modern” desire for few children and the desire to produce sons.

Finally, the conclusion situates this research back in the larger context of national

political instability and slow economic growth, of globalization, and of debates over cultural relativity, universal human rights, and applied anthropology. It offers a

discussion of the challenges of approaching maternal health in developing countries using

critically applied medical anthropology. I conclude by reemphasizing the limitations of

using culture or religion as an explanation for women’s positions and health behavior,

and raising questions about how to better address issues of local and global inequality

through anthropological research.

CHAPTER 1

METHODS AND ENCOUNTERS

The research setting

The dusty, noisy bus stop, the end stop on the bus route originating in

Kathmandu, has characteristics familiar to the more urban parts of Nepal. The chiya

stalls, one-room snack shacks of sorts with fried treats and “chow-chow” that double as

an informal, poor-man’s bar, and a variety of open-air shops offering all of the services

one might possibly need (though on a small scale) lined the street and the narrow

footpaths between buildings. A surprising range of goods could be produced upon request at even the smallest one-room hardware store or medical shop. The bazaar of

Vishnumati was flanked on one side by the row of busses and micros waiting to turn around and head back to Kathmandu with their passengers (mostly human, but an

occasional goat or sack of rice or vegetables was not uncommon). Much of the bazaar is

dominated by a Hindu temple and the stalls selling various religious artifacts necessary

for making an offering, serving mostly the out-of-town devotee who had not planned

ahead and brought the necessary items. Stalls filled with vivid flowers, fruit, uncooked

rice, multicolored string woven together to be tied around the neck, and fruit create a

colorful approach to the temple gates.

I lived across the street from the mandir (temple) in a three story concrete

building with the standard roof-top patio (kousi). Around the bazaar most of the

buildings follow this basic rectangular, concrete, multiple story architectural pattern. It is

common to see bundles of iron rods sticking out the top of a single or two-storied home –

a sign that the family couldn’t afford additional levels at that time, but plans to build

another floor. The flat, open patios on the roofs of the buildings were the location of

many daily activities such as drying chilies or wheat, hanging clothes to dry, or doing any activity that could be taken outside during the winter so that one could sit in the warmth of the sunlight. My adopted Nepali family, like most families in the area, didn’t bother with the portable heaters that were on the market at that time – so escaping the penetrating chill of indoors during the day was a great pleasure.

The kousi was also the perfect location for watching the world go by – a most appropriate phrase to describe observing the movement of people and goods from a house located on the only road leading down into the bazaar and toward Kathmandu from the upper regions of the Village Development Committee (VDC – an unincorporated rural area composed of 9 wards, less developed than a municipality). From this vantage point, both the temple grounds and the street were directly below. This meant that the music played by hired bands as part of wedding processions would beckon from the street below, especially during wedding season. All-night, and sometimes several days, of special musical praises to the gods (bhajan) were audible from the house. This unending singing/chanting, usually accompanied by an accordion, was projected by a loud speaker.

I dared wonder aloud once or twice if god, too, did not have to sleep? While having

one’s morning cup of chiya leaning on the concrete wall of the kousi, one could discuss with family members the parade of uniformed school children of all ages in their pleated skirts, slacks, and ties holding hands or teasing one another, men and women with wage

earning or salaried jobs hurrying to reach the office on time, and a handful of expatriate families in their sport utility vehicles avoiding potholes and pedestrians, as they all made

their way down to various destinations in the valley below.

Another fixture on the street below, though her presence was not as predictable as the flow of students and workers, was a homeless woman who, according to locals, had multiple personalities that spoke different languages. Her rants while standing on the

corner became part of the normal life of the street. Gangs of street dogs prowled the

street and alleys, looking for food and defending their territory against newcomers.

During the hot season before the monsoon rains began, whiffs of putrid air would

circulate around the bazaar from a combination of improperly disposed garbage and

human waste. In 2003 and 2004 there were open gutters lining the street through which

sewage also ran. (An underground sewage pipe was installed not long after I left in

2005.) The come and go of tourists on special holidays associated with the mandir, along

with community events on the temple grounds on special occasions, were also part of the

normal progression of weeks, months, and years. During the harvest season for wheat,

one or two enterprising families utilized the vehicle traffic on the paved road in front of

their house to separate the wheat from the chaff. Thus the street in this community is more a stage than a backdrop to many social interactions and daily practices.

The paved road around the mandir gives way to dirt as one heads up the mountain

side. However, with the rapid pace of development in this community, there is a good

chance that dirt road is being paved as I write. As one heads uphill, the houses become

more spread out, mud and brick houses of the traditional architectural style become more

plentiful, and agricultural fields and boulders begin to dominate the landscape. This

snapshot will not last long, as the buying and selling of land was one of the most profitable businesses available to local people. The real estate in this area was coveted by relatively wealthy Nepali and expatriate families alike – the land overlooked the

Kathmandu Valley, it was above the layer of smog that plagued the valley, and the nearby mountain ridge promised a lack of development on one side. In addition, the hot season was several degrees cooler than in the valley floor because of a combination of several factors, not excluding the altitude and the fact that the sun was partially blocked

by surrounding mountains for a portion of the day. The increase in the value of local land

resulted in some long-established local families selling their land for a substantial profit

and dramatic change in lifestyle, and in the irony of some impoverished families who

chose to continue their agricultural lifestyle in poverty on land worth more than they

would otherwise see in a lifetime.

The dirt road runs up the part of the mountain with the least intensity in slope, and

several side roads jut off on both sides. After a stretch of fairly flat land on either side of

the main road, the incline gradually becomes sharper. At the time of my research only

footpaths winded up these steep inclines. A walk to the bazaar and bus stop from some

of the most remote houses would take approximately 30-40 minutes in one direction.

These areas were not remarkably different from the remote, rural areas of Nepal, in which

roughly 80 percent of the country’s inhabitants live.

This covers the basic physical outline of the two wards in this study. The demographic composition of the population living in this VDC and its in-between status, neither a village nor a city, poised on the edge of the Kathmandu Valley with the urban metropolis below and the rural expanse beyond the ridge of mountains that encase the

valley, are what drew an anthropologist wanting to understand gender and reproductive

behavior among Hindu-caste Nepalis. Vishnumati’s location and semi-urban status made

it an ideal place to study social change; it was middle ground between the extremes of

Kathmandu and the village, both literally in terms of physical location on the outskirts of

the valley, and in terms of its combination of urban and rural characteristics. It was a

manageable microcosm of social change.

The other reason for choosing this site was the distribution of ethnic groups

represented in this population. Since this study investigates the interaction and changing

nature of social hierarchies, an ideal population for the purposes of this study was one in which the majority of people were part of the Hindu Parbatiya caste groups, including

Bahun (or Brahman), Chhetri, Thakuri, and the Dalit groups. The “Parbatiyas,” or

“people of the mountains,” is a moniker used more commonly by social scientists trying to establish and discuss cultural distinctions among the multitude of disputed ethnic and

caste groups in Nepal than by local people. The name designates a heterogeneous group

whose linguistic ancestry is traced to the Khasas, an Indo-European linguistic group who

migrated to present-day Nepal from the northwest (Whelpton 2005). The Khasas were

the original speakers of what became known as Nepali, which shares its Sanskrit roots

with Hindi and several other languages in the region. The Parbatiya group, then, is based

on linguistic and historical distinctions, but ultimately is a loose approximation of a

cultural group which is highly heterogeneous because of the subgroups that comprise it

and the porous, increasingly superfluous boundaries that demarcate it. Limiting the

present study to such a group is necessary to maintain some sense of cultural consensus.

The choice of this group allows me to utilize and speak to the literature on family,

gender, and health from northern India. Although there are historical and cultural differences – caste hierarchies differ in the specificity of groups and the rigidity of the system, and Nepal does not share India’s colonial history – Hindu-caste Nepalis do share much of a cultural and religious worldview with Hindu groups in northern India.

In the spring of 2003 I conducted preliminary background research on potential sites around the Kathmandu Valley. This included several trips to the Central Bureau of

Statistics in Thapathali in order to obtain detailed census data on the demographics of a few VDCs and municipalities. After charting the ethnic breakdown of a couple of sites,

Vishnumati emerged as an ideal site because of its large percentage of Hindu-caste residents. According to the latest census data (Central Bureau of Statistics 2002), the

Parbatiya castes (including Dalits) comprised almost 44 percent of the total population in this location. As can be seen in Figure 1.1, there is also a large number of Tamang and

Newar populations in the VDC; but neither of these groups dominate the local population and thus do not set the tone for the VDC or municipality, as Newari communities do in several locations in the valley. Figure 1.2 highlights the various castes (or, more appropriately, jaati) that make up the Nepali Parbatiya castes. Vishnumati seemed like the perfect site for the research topic because of its demographics and semi-urban characteristics.

Through acquaintances, I found a family in the chosen VDC with a daughter just a few years younger than me. Her family had supported her in finishing a master’s degree. This boded well for the kind of acceptance and space that I would need in order to do my research, but I had no idea at that time how generous and caring that family would turn out to be. After visiting the family for a few days, we agreed that I would

Figure 1.1

Ethnic and Caste Groups* in Vishnumati, 2001

3% 1%1%1% 1%1% 1% 1% 1% Chhetri 2% 28% Tamang 3% Newar 3% Brahman Magar Thakuri 4% Sarki Rai 5% Gurung Damai/Dholi Bhote Kami Unidentified Caste 8% Unidentified Dalit Tharu Sherpa Sonar 20% Other

16%

Figure 1.2

Parbatya Caste Groups Highlighted, Vishnumati, 2001

3% 1%1%1% 1%1% 1% 1% 1% Chhetri 2% 28% Tamang 3% Newar 3% Brahman Magar Thakuri 4% Sarki Rai 5% Gurung Damai/Dholi Bhote Kami Unidentified Caste 8% Unidentified Dalit Tharu Sherpa Sonar 20% Other

16%

return as a “paying guest” three or four months later after funding arrangements and

Institutional Review Board papers were all settled.

Enter the Anthropologist

I returned to Nepal the following fall, September of 2003, for 11 consecutive months of research with a 3 month follow-up visit in the spring of 2005. After confirming with the family that I was still welcome, I moved into a small room with the

two daughters in the family. I was grateful for this initial cramped arrangement, as it

eased my transition and helped me feel connected. Later I moved into a room of my own

on the third floor, after we furnished the newly finished bare concrete room with a small

piece of carpet, some curtains for the large, unscreened windows, and a curtain for the

doorway. My new room on the top floor was next to the family kitchen and a storage

room. This granted me much-needed privacy to work, yet I was still easily accessible to the family members and vice versa. I was even more grateful for this second arrangement.

I started out as a guest in the house, but over the weeks as we grew accustomed to one another many of the formalities fell away. Throughout my stay but especially in the beginning, I was conscientious to the point of being paranoid about not troubling the family members. I was constantly worried about doing something wrong, in spite of my cultural knowledge and prior experiences living in Nepal. I observed that even the smallest action did not go unnoticed by the family, and this only fed my apprehensions.

Despite my concerted efforts, I must have inconvenienced them hundreds of times

without ever knowing it. They never mentioned such things in my presence, and I was

thankful for their patience.

All Hindu families in the region recognize, if not adhere to, the pollution (juTho)

rules (sometimes referred to as ritual pollution in order to distinguish to English speakers

the difference from secular dirtiness or germs) that guide people’s interactions with

substances and with other people. The specificity of the rules and the level of adherence

to them depend upon caste status, according to my informants; but based on informal

observations I would speculate they also depend heavily upon level of education and the

extent to which families subscribe to a “modern” lifestyle. In my household, the family

fell somewhere in the upper range of observing pollution rules, but were not as strict, for

example, as the typical Bahun family in the community. I was very familiar with the

fundamental concepts and rules of the system: pollution could be transferred most easily

through touch or contact with a person, body part, or substance in a polluted state.1 Two substances stand out as being more powerful potential transmitters of pollution: water and cooked rice (bhaat). Someone unfamiliar with the cultural context would need to know that these two substances are the most basic forms of sustenance – fundamental in terms of survival – and because of their centrality to life, carry much symbolic meaning. Rice is eaten with every meal; indeed, its importance to eating and feeling full and content is often talked about explicitly. It is evidenced linguistically by the fact that the word

“bhaat” (cooked rice) is used to refer to a meal in a general sense in addition to cooked rice. Another word, “khaana,” works in the reverse direction: “khaana” means food, but is also used to refer specifically to cooked rice or, more generally, a meal with cooked rice. The exceptions mentioned earlier notwithstanding, someone from a higher caste

1 See Cameron 1998 for more detailed explanation.

may refuse to accept drinking water or a meal of cooked rice from someone of a lower

caste.

How these general rules translated into detailed behavior in the family kitchen, an

area of high risk for polluting food or water consumed by others, became an ongoing

lesson in how the underlying logic of the belief system played out in everyday details. In

one corner of the marble floor kitchen was an area designated for washing dishes, pots,

and pans. It consisted of a faucet protruding from the wall, and a raised area with a lip

and a drain built up a few inches from the floor. It was explained to me that this area for

washing dishes was juTho (ritually polluted), and to be careful in not contaminating food

or cooking vessels by contact with this area. Despite this area being cleaned after the

metal plates and bowls were washed, it remained in a general juTho state. (The dishes were juTho from the hands that had touched them as people ate and their fingers came into contact successively with their mouths and plates.) The floor surrounding this one corner of the kitchen, however, was an essential work surface in food preparation. The floor was a common place to sit and cut vegetables, the place where the mortar and pestle stayed and was used, and where cooking would take place any time an extra source of heat was needed in addition to the two gas burners stationed on the counter.

Learning to squat or sit on the floor for extended periods was part of the phenomenological experience of becoming more Nepali. I slowly began contributing to

food preparation by cutting vegetables and cleaning up after meals, but the family

protested loudly when I would attempt to wash their dishes. Occasionally one of the

daughters would even push and pull me away from the sink in protest. The progress that

I made at being allowed to wash the dishes became a symbol of the extent to which they

considered me part of the family. I was intent on leveling our respective positions as

much as possible, especially since I was mostly unable to do so in interactions outside of

the family. Once I was allowed to wash dishes with only weak protests from the others, I considered this a major triumph in overall acceptance. I was not as successful at developing the back strength to bend over and wash the dishes – and a family member

would often slip a small stool under me as I worked to allow me to sit. I had to hand

wash my clothing in buckets in the same fashion, using a stool instead of squatting. Such

matters, along with eating with one’s hand, giving up the use of toilet tissue, bathing from

a bucket of water, may seem trivial – but they are key distinctions in one’s “being in the

world,” and may have health related implications. The anthropological tradition of living

like locals provides insights that would otherwise be inaccessible. Some things must be

experienced in order to be understood.

A few times when the other family members were all called away from the house

for some business at the same time, I was even allowed to cook for the entire family.

With their own best interests at heart, they tried to avoid that scenario as much as

possible. My skills at making daal bhaat were not comparable to theirs, despite watching and helping numerous times. I would argue that this is not completely my fault,

considering that the mother and two daughters were exceptional cooks and produced

some of the best daal bhaat I had ever tasted. My versions of pad thai and vegetable

chow mein were passable dishes for khaaja, the midday snack, and I thus passed muster

of at least being trustworthy to fix snack for everyone. I was also deemed worthy of

making chiya, which involves the perfect mix of loose tea, water, cow milk, and sugar

(sometimes with a bit of black pepper or ginger for their warming properties).

Living with a family was immensely important for the success of my research and my personal well-being. My status as a member of a local Nepali family was a

prerequisite to gaining credibility in the eyes of families in the community and the ability

to conduct research there. Belonging to a family helped distinguish me from

international non-governmental organization (INGO) workers or family planning

advocates who had conducted surveys in the area in the past. I quickly learned that mentioning where I lived locally helped put people at ease, and as a result I tried to work that fact into conversations. I profited also from the reputation of the family with which I

lived as a hard-working, upstanding family in the community. Being part of a family

meant access to others, but at the same time it brought the responsibility of not damaging

their reputation by acting too much like a Westerner or in a disrespectful manner. As in

small towns around the world, people in the community were aware of my presence and

movements. I responded by dressing conservatively, often in the common dress worn by

young women in much of South Asia, and making sure that I limited leaving the house or

returning home after dark. Learning the rules of my status as an honorary daughter in a

family, and gradually as a member of the community, was invaluable to developing an

ability to discern the slight yet significant acts of deference and authority that imbue

every social interaction.

Overview of Methods

While becoming part of a local family is not a method per se, it certainly affords a

great deal of insight into the lives of the people with whom one is working (in addition to

the advantages mentioned previously). Participant observation with “my” family served

52 as a profound source of knowledge and understanding that could only be obtained through intimate relationships in this somewhat conservative, private culture, and that could be used to understand and contextualize other people’s behavior. I also had the advantage of directly asking them to help explain certain actions, beliefs, or references encountered in the community that I did not understand. I made it clear to my family, however, that their lives were not the subject of my research – I only used my experiences with them as a contextual backdrop for understanding the behaviors, attitudes, and events that I was studying. Thus my family’s personal stories are not told in this work.

Participant observation has long been touted as one of the strengths of the anthropological approach to research, and versions of it are becoming more recognized among the other social sciences. Qualitative research has experienced a recent surge in popularity outside of anthropology in sociology and, to a limited extent, in demography.

Conducting intensive research contextualizes and gives meaning to the information gained through surveys and interviews (Knodel 1997). By checking actual behavior against reported behavior and by establishing rapport with respondents, participant observation can improve validity. For example, watching family communication and decision-making processes in the home, or observing different family members’ roles in daily activities, serve as important cross-checks to what people report. Participant observation is also advantageous because it enables one to better detect potential problems with research design early in the study, and it guides appropriate adjustments in the design at various points throughout the project (Axinn et al 1991).

Participant observation has the advantage of incorporating spontaneity and

serendipity into the research process; it “brings the researcher into confrontation with

social life as it unfolds and plays out on terms mostly of its own making” (Krause 2005).

Krause explained with a metaphor that will help those not involved in the ethnographic

project become more familiar with the process: “Just as improvisation in musical genres

such as jazz or bluegrass builds on a foundation of competence, practice and sensibility,

so, too, does the art of ethnography” (2005:594). Great musicians make improvisation

seem easy; and great ethnographers make techniques such as participant observation

seem simple.

Grounded theory was designed along similar principles. It maximizes the

flexibility of projects goals and research design in order to allow one to respond to what is going in a specific context at the ground level at numerous points throughout the process of conducting research (Glasner and Strauss 1967, Strauss and Corbin 1990). I incorporated grounded theory into my methodology in order to ensure that my categories and questions were not imposed upon the people I was researching without reflection; through listening to and reflecting upon women’s own concerns and trains of thought throughout the fieldwork, I was able to pick up on themes that would have otherwise gone unnoticed. Some of the most productive moments of my research occurred when conversations unfolded naturally out of interview questions. As a result, the final product thus represents a meeting of my categories and interests and theirs.

Through immersing myself in the community by becoming an “adopted” member of a local family, I effectively was conducting participant observation with almost every interaction. I had to relearn how to do some of the most basic daily activities, and rarely

was I ever so involved in the moment that I escaped my self-awareness as being an observer. As I became more of a member in the community, I attended weddings, first rice ceremonies, and special observances at the temple. I traveled with one devout

woman to listen to her guru speak, and I occasionally visited the Tibetan monastery and

the Hare Krishna center in the community. Negotiating public transportation, strikes and protests, and even the purchase of vegetables were all exercises in mastering cultural values and norms.

One of the first steps in initiating the research project was meeting with the local

leaders at the VDC office. In addition to wanting them to be aware of my identity and the reason for my presence, I hoped to obtain a map of the VDC broken down into wards

(the smaller political subdivisions). After several visits and much waiting, I came to

discover that the maps in their possession were dramatically too large in scale and not

useful in gaining an overview of the VDC. From the one small, hand-sketched map of

the VDC that they did have, along with some clarification questions, I was able to

determine the boundaries of the two most populated and central wards in the VDC. One

was centered around the bazaar, and the other essentially followed the main road up the

mountain. I chose the population of these two wards as the focus of my study.

Initial Survey

Shortly after my arrival, I began to map all of the households of the two wards. I

recruited the son of my family as a local expert whose status as a male, whose bold yet amiable and disarming manner, and whose motorcycle helped me accomplish this formidable goal in less than two months. In each building I determined two things: the

number of households living there and the number of members in each household. This

was done painstakingly by asking the question several different ways with slightly altered

wording in order to ensure people understood that we were defining “household” as a

group of people who live and eat together from one kitchen or stove. For instance, if two

groups of kin lived under the same roof but cooked and ate separately, then this

distinguished them as two separate households. This definition of a family unit or

household was derived from local understandings, but there was no singular word that

distinguished this kind family unit from general, inclusive terms for family or home. In

November of 2003, I finished enumerating all 794 households in the two wards.2 Thus households, as defined here, became my unit of analysis for the initial survey.

I drew a random sample of 250 households (using a random numbers table) from the total of 794 in order to survey them for basic demographic, household history, and birth information. This initial survey was necessary at the outset of the research project in order to discover the most salient categories of household composition in this cultural context and to establish the prevalence of common practices surrounding fertility and birth. I had hypothesized that living in a nuclear versus extended family household would have significant implications for maternal health, but first I needed to find out if these two categories were culturally relevant, especially when taking the domestic cycle into account. I wanted to avoid imposing outside categories onto the cultural reality of this group of people.

2 This included a handful of makeshift squatter shacks. Other than the one mentally ill woman, I never encountered a homeless person. I say this with reasonable confidence as I covered practically every square foot of the two wards on foot hundreds of times in the fourteen months. Of the total number of households, nineteen were single men living alone, and six were elderly women living alone. I excluded the monastery and two boarding schools that fell within the two wards in the initial household enumeration. At least five new buildings were completed during my research, but new buildings or households were not added to the study.

The survey also provided many useful basic demographic characteristics of the

population, such as age, education level, and occupation of all household members. At

the end of the survey I included ten questions on birth including the location (at home or

in a hospital) and who was present, any health complications, and immunizations. The survey was designed so that it would take a short amount of time to administer, between

15-30 minutes, and so that there were no controversial questions or subject matter.3 I

tested drafts of the survey using acquaintances, made improvements, and then field tested

the survey using two research assistants and random community members. In the end I

had to be present at all but a handful of the surveys, for we discovered that without my

presence some people tended to quickly dismiss the young men. Thus identity politics

came into play immediately in the research project. A total of 248 questionnaires were

administered after two households refused.

One of the strengths of this project is the incorporation of quantitative data along

with detailed, thick description (Geertz 1973) of women’s lives and social relations in the

family and community. The quantitative data help to balance the specificity of the case

studies and provide a way to judge how common or unique the information gathered from the case studies was. The other major role of the survey was to provide a sampling frame for the case studies.

Case Studies

The data gathered in the survey provided information necessary for choosing families for in-depth case studies according to a sampling matrix of the following

3 One of the selected households consisted of a woman with Japanese citizenship and her Nepali husband, so I threw out that selection and replaced it with the next random number.

characteristics: family type, caste, socioeconomic status, education, and age. This was

necessary to capture and account for variation created by each of these categories. This

endeavor was, in itself, instructive: a household representing every combination of these

variables was not available because several of the categories were correlated. For

example, there were no low-caste families in my sample that were in the category of

moderate or high socioeconomic status. Being a member of a low-caste group was still

correlated with low socioeconomic status in the area. Moreover, there were surprisingly

few low-caste families in the sample at all, despite my efforts to ensure that impoverished

families were not excluded in the original household enumeration. It took several months

of inquiring why this might be the case before someone admitted that many low-caste

families lived nearby but in a different ward from the two included in my study. Several

low-caste families were completely integrated in the bazaar within my study area, but

there was also a vague settlement of land-owning low-caste families in a more rural area

bordering the two wards in my study. In order to include several low-caste families in

my case studies, I had to seek out families in the two wards that were not part of my

original household survey. The only two low-caste families in the study area that were

joint families (and that fell into the middle SES) refused to participate in the study; so all of the low-caste (Dalit) case studies were nuclear families.

Ultimately I selected 30 case studies that represented de facto joint and nuclear families of each caste and, other than the low-caste families, of middle or low socioeconomic status. Wealthy Nepali families were few and anomalous in the area, so I excluded them from my case studies. As explained previously, I limited my case study households to Hindu-caste Nepalis – a substantial and influential group, yet only one of

many diverse cultural groups found in Nepal. One family experiencing fertility problems

dropped out of the study after they became convinced that I could not help them in that

regard, and two families moved.4 Thus I had a total of 27 case studies (see Table 1). I

interviewed the main married women of each household using a semi-structured, open-

ended format an average of five times over the final seven months of my research

(depending on how often we were interrupted). The first two introductory rounds of

household interviews were not taped, but after building rapport I used an audio tape

recorder for the remaining three interviews. I also had many informal conversations with

various case study family members along the way. Informal group banter and the

interjections of family friends were some of the richest sources of cultural information.

Table 1.1 Number of Case Studies by Caste and Family Type

Extended Family Nuclear Family TOTAL Bahun 3 3 6 Chhetri 4 3 7 Thakuri 4 4 8 Dalit 0 6* 6 *One of these families was unusual in its composition: it included a daughter who left her husband and returned to her parents’ household with her young daughter.

In addition to the initial survey, participant observation, and case studies, I

conducted interviews and observed prenatal checkups and counseling sessions at local

family planning clinics and in the maternity wards of two major hospitals in the

Kathmandu Valley. The first, casually called “Teaching” because it is a teaching

hospital, was less than 15 minutes away by public transportation and even faster by taxi

or ambulance. The other hospital is a significant drive (and thus prohibitively costly),

4 I spent a month trying to locate them, to no avail. It was several months into the research period at that point, so in the interest of time I could not replace any of the three dropouts.

between 30- 45 minutes in a taxi depending on traffic, and difficult if not impossible for a

woman in labor to reach by public transportation. Because of its good reputation, a few

women in the area did use the Thapathali Maternity Hospital for prenatal care. I also

interviewed the staff and observed a prenatal check-up at the nearest sub-health post,

about a fifteen minute walk from Vishnumati or a brief bus ride. And last, I interviewed

the director of a local family planning clinic located in the bazaar. This one-room clinic

provided temporary contraception and counseling, but no other services.

Encounters

Doing ethnography in urban or semi-urban area(s) creates a different set of

politics: the ethnographer is no novelty like in rural, remote areas. The anthropologist

might be perceived instead as simply intrusive or bothersome, as yet another tiresome

NGO worker, or, depending on their appearance and identity, a general embodiment of imperialism. On the other hand, people may welcome the ethnographer with genuine kindness and sense of interest, enjoy swapping stories, or cooperate with hopes of gaining something in the process. I experienced all of these scenarios, luckily with the positive receptions far outweighing the negative.

My encounters included interviews with a mother with three young children who suffered from depression or alcoholism or both, a father whose kindness was so remarkable that his wife and two daughters could not stop praising him, and several women who were abused by their husbands. This last brand of encounters convinced me to abandon a section of my research in which I intended to interview husbands about what it means to be a good son, husband, and father, and their hopes for offspring and

future living arrangements. Given the fact that wife beating and, in some cases battery,

were common, I determined that interviewing men would put women at further risk. This

was a disappointment in terms of the study, but it was the obvious choice. In addition,

conducting interviews about things that men probably would not discuss openly – things like infidelity, alcoholism, and abuse that are part of life, but that go unacknowledged publicly – seemed futile.

Doing ethnography is contingent upon gaining access and maintaining good relationships and impressions, for the ethnographer is on shaky ground. She does not offer much to the participants in the research directly, and she asks much in terms of people’s time and patience for answering what sometimes seem like obvious questions, or detached, philosophical questions that require a level of conjecture unfamiliar (or impractical) to the individual. My gifts of fruit and small trinkets to the case study families were more a token of thanks than a form of remuneration. By the end of the study, most families had formed a more informal relationship with me, involving much joking, teasing, and gossip. Two women, though, had decided they had nothing to gain from me and therefore quit the study. Thus in participating in the ethnographic project, both sides were bringing certain expectations and requirements to the table; and these varied according to the individual and evolved over the progression of time and exposure to one another.

As detailed in the introduction, ethnographic research is far more than just qualitative “data collection.” For better or worse, it involves the repeated meeting of two lives – subjective real life encounters. These encounters are fraught with power imbalances and identity politics, but at the same time may function as a collaboration

towards lessening a social ill or increasing the level of understanding of one’s own

society and its interconnectedness with others and globalization. The end product, too, is richer for this: the ethnographic text is hardly a collection of “just facts” (Abu-Lughod

1993). It aims to give insight into the very nature of what it means to be human.

CHAPTER 2

SOCIAL HIERARCHIES IN CONTEXT

In order to situate the people’s lives portrayed in the following chapters within the meaningful structures of society, the relevant social hierarchies for Hindu-caste Nepalis will be considered in this chapter. I was often told that caste no longer mattered in Nepal, as discrimination on the basis of caste had been outlawed. Caste stereotypes, no longer bolstered by public and political support, have been relegated to the insidious realm of privately held, conservative opinions about “us” and “them.” This is not to deny the importance of ritual status and the legacy of discrimination that continues to shape people’s lives, but to point out a growing alternative measure of success, and also stratification, whose implications need to be examined at the micro level – class.

In addition to considering the role of class, a call has been made for a gendered approach to studying caste (Fruzzetti 1998). Historically, the experience of caste for women has been distinct from that of their male counterparts because of a surprising interaction between gender disparity and caste inequality: greater autonomy for low-caste women than high-caste women. I use the case of Nepali Hindu-caste women to examine the significance of caste in their lives, and to complicate the previous supposition along two lines: the role of socioeconomic status in determining women’s autonomy, and the extent to which “autonomy” captures low-caste women’s positions. More broadly, the findings imply that the relationship between caste and class is slowly unraveling in semi-

urban and urban areas of Nepal, thereby verifying the inadequacy of previous models that

privileged ascribed ritual status and adding to the research that focuses on person-

centered, local constructions of hierarchies such as caste, class, and gender.

Constructs of caste in South Asia

Since caste became an object of inquiry of social scientists in the early twentieth century, changes in concepts of caste have reflected the flow of prevailing theories in sociology and anthropology over the years, and also the theoretical divides. The rift between structural-functionalist, interpretivist, and materialist paradigms created a debate over the meaning and the model of caste in which various sides would never agree, given that they were coming from fundamentally different perspectives and striving toward different goals. In the wake of far-reaching contributions from Bourdieu and Foucault, more recent theoretical trends have shifted the focus from creating a unifying model of caste towards more experientially defined, practice-oriented, power-conscious versions of caste. A reified model of caste has been replaced by fragmented accounts taking into account variation, historical circumstances, and the dialectic between culture and agency.

Along similar lines, religious studies scholars have rejected monolithic interpretations of

Hinduism.1 Although the “classical” theory of hierarchy has been discarded by many

contemporary social scientists on both theoretical and historical grounds, the legacy of

caste remains and needs to be accounted for.

Early theories of caste were based primarily on Brahmanical texts in the tradition

of armchair anthropology. David Mandelbaum’s Society in India (1970) was one of the

first comprehensive works based on empirical research and ethnography. Louis

1 See the collection of articles in Journal of the American Academy of Religion, Vol. 68, No. 4, 2000.

Dumont’s approach to the study of Indian society was a mixture of the two: he included references to ethnographic studies of the day, but he was most concerned with developing a totalizing model of a pan-India caste system based on early Hindu texts. His approach marked a departure from the prevailing preoccupation with functionalist explanations; he was interested in people’s ideas in addition to their material culture and social institutions. Being trained in the French sociological tradition, Dumont stressed the role of ideology in shaping behavior. Some scholars believed that creating any unified model of caste was impossible, so unsurprisingly they found faults in the model Dumont created. For some, this was more a rejection of what would come to be known as structuralism rather than the model itself. T. N. Madan (1971) pointed out that models are neither real nor are they intended to be exact replicas of reality; rather they intentionally overcome the complexity of reality by introducing a particular kind or order and simplicity to it. But Madan also criticized Dumont’s model for not allowing for change. Dumont’s structuralist approach to society had allowed him to create, with no misgivings, a synchronic model of caste.

Dumont’s Homo Hierarchicus (1966) generated much response. His contemporaries working in the area felt obliged to respond to such a seminal and monumental work, and scholars concerned with South Asia still use his Homo

Hierarchicus as a point of departure. Dumont was lauded for outlining fully and extensively the fundamental opposition between the pure and impure as the foundation of the caste system, the underlying principle linking Cèlestin Bouglé’s three relational properties of separation, hierarchy, and interdependence (1971). His declarations that the caste system as an Indian institution had its full coherence in the Hindu environment, and

that caste was present only where the disjunction between status and power was present,

challenged scholars to articulate the differences among, and even within, local traditions.

Dumont was criticized for his focus on purity (Das 1977) rather than sacredness (van

Furer-Haimendorf 1971) and his conception of hierarchy (Kolendra 1976). Several

scholars also challenged him on the fit of his model with Indian facts using their first-

hand experiences of field research, a few of whom are worth reviewing here.

In a special 1971 issue of Contributions to Indian Sociology devoted to

responding to Dumont, Gerald Berreman pointed out that Dumont had privileged a

Brahman perspective and therefore his resulting model of caste represented a limited,

one-sided Brahman worldview of “…what the caste system of Hindu India ought to be,

according to those who value it positively: it conforms well to the theory of caste

purveyed by learned Brahmanical tracts” (1971:23). Berreman held that if Dumont had

talked with low caste people he would have uncovered a different version from the

idealized one laid out in Homo Hierarchicus. Other drawbacks to privileging a Brahman perspective were postulated by Dirks, who contended that the secular power of the

Kshatriya kings was not encompassed by the ritual hierarchy of the Brahmans (1987),2 and by numerous scholars who addressed the issue of low-caste members’ acceptance or denial of their assigned status.3

Andre Beteille disagreed with Dumont for reasons similar to those of Berreman,

protesting an overly intellectual interpretation of caste that privileged ritual status and

religious values above all other interests. Beteille, based on his field research in Tanjore

District in South India, had already concluded that caste was not the sole organizing

2 Also see Quigley 1993 for another analysis that focuses on the role of kingship. 3 See Inden 1990 for a discussion of the orientalist view of caste and Indian agency; for some of the contributions of subaltern studies toward establishing Indian agency see Guha and Spivak 1988.

66 principle of Indian society (1965) by the time Homo Hierarchicus was published. His focus on social and economic relations instead of ritual led him to emphasize the role of class over caste in structuring and defining social relations. He wrote,

The hierarchies of caste, class, and power in the village overlap to some extent, but also cut across. It is the argument of this study that in the traditional structure the cleavages of caste, class, and power tended much more than today to run along the same grooves. The Brahmins were the landowners, and they also constituted the traditional elite. This is no longer the case at present. The social system has acquired a much more complex and dynamic character, and now there is a tendency for cleavages to cut across one another (1965:4-5).

He observed that many areas of life were becoming “caste-free”; land ownership, occupation, and education were not to the same extent dependent upon caste. New institutions had been created, and with them new bases of power. The class system was gradually disassociating itself from the caste structure. Betielle also recognized the limits of class and ownership for social honor. To improve one’s status in the caste system one had to concentrate one’s efforts within the ritual domain; improving one’s class status was not sufficient for social mobility. Moreover, he concluded that caste, class, and power, while useful as analytical tools, cannot be treated separately except at the most abstract level. Beteille’s recognition of the dynamic character of the social system is notable because it came before the predominant theoretical climate had turned to the more fluid and diachronic theories of culture that have gained popularity in the past three decades.

Interpretivists have defined caste as a unique symbolic world tied to Hinduism that could not be cast as a type of social structure to be used in comparative studies. The extent of their agreement stops there. Although heavily influenced by structuralism,

Dumont exhibited interpretivist leanings in one of his fundamental tenets: that caste has

its full coherence in the Hindu Indian context. While interpretivists agree that an

indigenous understanding is necessary for caste or any cultural matter, McKim Marriott

took the approach further than others. He argued for an indigenous sociology (1990), meaning that Indians should be the ones to conduct Indian sociology.

Others were satisfied with a more moderate approach to correcting the

essentialization of India as caste. This involved allowing indigenous people to speak

freely and working to understand their categories and cultural logic through language,

behavior, and ritual, without imposing the researcher’s own preconceived categories in

the process. Fruzzetti and Ostor (1976) argued for a new kind of structural comparison, one founded on indigenous categories – the cultural construction of person. But at the

same time, they rejected “…the reification of an Indian uniqueness that would make of

India another planet” (1984:4). Their approach, stressing the indigenous ordering of

relationships, was influenced by the innovations Schneider (1974) had made in the study

of kinship. They aimed to dissolve the concentration of the field of anthropology on

caste and kinship terminology. They sought to bridge the comparative versus interpretive argument by acknowledging the “play of universals and indigenous culture…” holding onto “the tension of a dialectic that constitutes the comparative and interpretive science of anthropology” (4). They continued, “Anthropology is concerned with analysis as well

as understanding, with universals as well as particulars” (4). They made explicit the

question: what do scholars mean by caste?

Contemporary studies of modern society in South Asia have tended to reply with

a myriad of local answers rather than attempts at constructing a revised monolithic

answer. “Caste cannot possibly have the same meaning or the same legitimacy for all the

68 individual members of such a large and diverse population” (Beteille 1996:152). I tend to believe that this was always the case, but that the modernization and urbanization that has occurred throughout South Asia in the past century is responsible for making such diversity more pronounced. Sharma contemplated, “…perhaps caste is only ever what those who practise it say it is at any given time, a fluid and variable category with little stable content… a historical label attached to a constantly shifting and fluid social arrangements…” (1999:60). Perhaps there is nothing “essential” about it. As for modern

India, Gupta claimed, “From politics, to gender, to economic interaction, the pure hierarchy is not just being questioned; it is, more often than not, left unattended by the wayside” (2004:vi). And Sharma admonished scholars not to allow discussions of social change to rely implicitly on the original principles of “the caste system,” or the

“traditional.”

Such statements do not imply that caste as it figures into people’s everyday lives is unimportant. It seems that it has moved from being the dominant, overbearing organizer of social relations to a suppressed rationale about how people are that comes out behind closed doors. According to Fuller (1996) for example, caste is delegitimized publicly by many Indians, and is not present in public discourse as a justified mode of action or thought. Likewise, Indians may tell anthropologists that they do not practice caste any more, that there is no caste now (Beteille 1996). Yet caste remains a potential source of pride, discrimination, or political maneuvering. Nepali people of privileged status often told me there was no practice of caste anymore, and I believed them until I started living as a daughter in a Nepali family and as a member of the community. Caste began showing up all around me. Looking behind public statements to discern the

69 normative understandings expressed in private discourses and everyday interactions revealed the significance of caste in people’s lives.

The distinctiveness of caste in Nepal

Scholars have not been tempted to talk about a caste system in Nepal in terms of a single system of hierarchy rooted in Hindu ideology. Caste as it exists in Nepal resulted from specific historical circumstances including the settlement of various cultural and linguistic groups and the eventual political unification of those groups. Those with political and cultural hegemony created a national language and identity, Nepali, and assigned all groups a status along a single model of hierarchy. This was met with various levels of acceptance and rejection. Since the abolition of discrimination based on caste and the advent of a more democratic form of governance, people of “ethnic” and low caste status have attempted to utilize their subjugated status to their advantage politically.

And all of the statuses are being re/created in terms of modernity and cultural heritage.

In contemporary Nepal, the fluidity, contestation, and arbitrariness of caste (jaat) categories are recognized by Nepali citizens; but this does not diminish the latent power of the categories.

Partly for the sake of simplicity and partly because of a dearth of historical and archaeological evidence regarding the migration and settling of various groups, descriptions of the history of indigenous peoples in Nepal often oversimplify the picture in one of two directions (Levine 1987). The first model of ethnicity in Nepal is one of unlimited ethnic diversity, and indeed, one is confronted at every turn by variation,

dissimilarity, and uniqueness. The cultural and linguistic diversity is staggering: it is

estimated that there are over 70, potentially over 100, distinct languages4 and cultural

groups.

On the other end of the spectrum, the second and opposing model of ethnicity

presents a limited set of ethnic contrasts. The set of groupings that grouped ethnicities

into a set of limited categories are composed of opposing sets of descriptors include

Hindu versus Buddhist, tribe versus caste, and mountain versus hill versus Terai plains.5

Several researchers have corrected the problem of a limited set of ethnic contrasts

through their analyses of ethnicity among the Tamang, Newar, and Thakali. As more and

more research was conducted in locations throughout Nepal, scholars noted that one

ethnic name was being used to designate disparate groups (Holmberg 1989), and that

people were strategizing their ethnicity, sometimes acting as a group and not just

individually, in ways to help them get ahead (Fisher 2001).

Guneratne (2002) postulated that status difference has begun to be explained in

terms of ideas about modernity instead of caste hierarchies, and he demonstrated this in

the case of Tharus in the Chitwan District of Nepal. He explains how the rhetoric of

“forwardness” and “backwardness” is employed to describe a set of attributes such as

level of education and success in business that are unrelated to any inherent

characteristics of the group. This distinction is crucial since it allows a complete divorce

from ritual attributes and redirects attention toward social, economic, and political

4 Sources disagree wildly on the number of distinct languages spoken in Nepal. The 1981 census cites 18, the 1991 census cites 32. The 2002 census cites a record 92 languages, in part because of a change in the recording of the response to the question, “Which language do you speak as a mother tongue?” Enumerators recorded the actual response of the person without using a list of languages and dialects for reference. It has also become more acceptable to report one’s language or identity as something other than the national language and identity of Nepali. Some linguists had previously identified 70 languages (Malla 1989; Toba 1992). See Yadava (2003) for further explanation. 5 See Levine 1987 for further discussion.

71 disadvantages that people have faced because of membership in a certain ethnic group under a specific set of historical circumstances. Backwardness thus “may be rectified by partaking in opportunities that are open to all in a modern, more or less democratic state”

(2002:84). “Backwardness” is a condition that is faultless and that may be overcome.

The state of present-day groups in Nepal may be traced to the historical movement of people into the areas which now constitute the nation. Unlike India, Nepal was never colonized. The codification of caste came not from a Western imperial power, but from a more local brand of conquerors and rulers. It began in the second half of the

18th century when Prithivi Narayan Shah, the king of the small state of Gorkha west of

Kathmandu, conquered the Kathmandu Valley after over two decades of sieges and blockades. From there, he and his successors proceeded to take control of a long stretch of the Himalayan foothills, forming the Gorkhali kingdom and the beginnings of the modern state of Nepal. After the initial period of rule by the Shahs (1769-1846), the

Ranas, a clan of the Chhetri caste, began to define the area as a nation-state called Nepal.

It was under Rana rule (1846-1951) that the caste system in Nepal was created (Gellner

1997).

In 1854, the first Rana ruler wrote the Muluki Ain in an attempt to consolidate previously divergent, local rules under a single corpus of laws backed by the state. In terms of social structure, the legal code organized society along the lines of a single hierarchy based upon Hindu principles. With this document, the state effectively arranged a multiethnic and multi-caste (Newars and Tarai Hindus have their own internal caste systems) society into a single ranking based upon Hindu notions of purity and

72 pollution. Ascribed status was the central organizing concept in this process (Hofer

1979). Attributes such as drinking alcohol, eating meat, or engaging in certain professions were used in the decision to cram an immense diversity of ethnic groups into a limited number of categories in the middle levels of the Hindu caste system – below the high-caste Parbatiya, Indian, and Newar castes, but above the so-called “untouchable” castes. The end result was a state-sponsored caste system unique to Nepal, particularly in its incorporation of ethnic groups (see Table 2.1).

Table 2.1 Caste Hierarchy of the Muluki Ain

“Wearers of the sacred thread” (taagaadhari) Upadhyay Brahman Rajput (Thakuri) Jaisi Brahman Chhetri Newar Brahman Indian Brahman Ascetic sects Various Newar castes

“Non-enslaveable alcohol drinkers” (namasinyaa matwaali) Magar Gurung Sunuwar Some other Newar castes

“Enslaveable alcohol drinkers” (masinyaa matwaali) Bhote (people of Tibetan origin) Chepang Kumal (potters) Tharu Gharti (descendants of freed slaves)

“Impure but touchable” (paani nachalnyaa choi chito haalnu naparnyaa) Kasai (Newar butchers) Kusle (Newar musicians) Hindu Dhobi (Newar washermen) Musulman Mlecch (Europeans)

“Untouchable” (paani nachalnyaa choi chito haalnu parcha) Kami (blacksmiths) and Sarki (tanners) Damai (tailors and musicians) Gaine (minstrels) Badi (musicians and prostitutes) Cyame (Newar scavengers)

The creation of this caste system was part of a much larger attempt to create a

unified nation-state and national identity. The hierarchy was designed, of course, from

the perspective of those in power – the Parbatiya Hindu high-caste groups. The extent to

which this legislation was followed is difficult to determine. James Fisher surmised, “As

official law governing such matters as caste hierarchy and inter-caste relations, it does not necessarily record what really happened, but it certainly indicates the Nepali government’s strict Hindu hopes for its people” (as cited in Guneratne 2002:74).

Guneratne (2002) pointed out that although the Muluki Ain had substantial influence in the areas tightly governed by the state such as the Middle Hills and the Tarai, the legal code probably had much less influence on the social relations and thinking of populations living in remote, inaccessible regions of Nepal. Nepal’s extreme topography and resultant remote areas make it difficult to govern – whether by the Ranas or by a monarchy fighting the Maoist uprising. King Mahendra finally replaced the revised

Muluki Ain with a new legal code in 1963. The new code did not support the caste system, but also did not explicitly abolish it (Hofer 1979). The modern state of Nepal does not recognize the institution of caste, and the law prohibits discrimination based on caste or ethnicity.

The groups codified in the Muluki Ain, called “jaat” in Nepali, are often glossed as “castes” in English translations, but that overlooks the colonial etymology of the word

“caste” and the historically and politically situated connotations and employment of that word. “Jaat” can be translated as “species,” “kind,” or “group,” and is used to refer to a wider range of groups than merely caste distinctions. Thus as Cameron observed (1998), jaat can sometimes designate difference without implying hierarchy. In order to maintain

74 the wider definition of this word, I use jaat instead of caste when referring to the distinctive groups outlined in the Muluki Ain. The ways that the Parbatiya inhabitants of

Vishnumati identified themselves were in three broad high-caste groups – Brahman or

Bahun, Thakuri, and Chhetri – and a multitude of low-caste groups. Overall, the most common distinction among Parbatiya jaat that I heard local villagers making was between small (saano jaat) and big (Thulo jaat). And while I listened to a few Thakuri and Brahmin men draw a small number of fine distinctions between the practices of their respective groups in behavior and speech, most people spoke in more general terms of saano or Thulo, which I render as low and high. I did not dispense with jaat categories in the research design because I did not want to gloss over distinctions that might be present, but in the end, it seemed that only high versus low mattered in most everyday discussions and interactions.

Although caste is no longer a system of social organization backed by the state, its remnants continue to influence social behavior. Guneratne characterizes this transformation as one that moves from hierarchical caste relations to relations amongst various ethnic groups with a certain degree of political equality under a constitutional monarchy. “When caste is abolished or delegitimized, it does not disappear; instead, in responding to new ideologies and practices of rule, it transforms into new social formations” (Guneratne 2002:79). “…[C]aste systems cease to be systems and decompose into their constituent ethnic components, whose mutual relations are organized along different bases of political competition and underpinned by different ideologies” (Guneratne 2002:82). Caste may cease to be a coherent system (if, indeed, it

ever actually was one on the ground), but the influence of caste on social relations does not disappear overnight.

In Vishnumati, ritual status was still valued by those at the top of the hierarchy to an extent (depending on individual characteristics and variation). Upholding Brahman restrictions on eating, hygiene, and worship were accompanied by a sense of ritual superiority amongst some Brahmans, whereas these prohibitions were dismissed by low- caste families as tedious and unnecessary. However, Brahmin priests continued to play a significant social role: they were called upon for certain ritual functions such as performing marriages, coming of age ceremonies, and death rites by families of high and low caste.

Certainly many aspects of urbanization preclude an adherence to rules of untouchability – whether one is concerned with people in temporary states of pollution such as menstruating women or with Dalits. Consider, for example, a public transportation system that packs passengers inside vehicles in a way that, from an

American’s perspective, violates proxemics and creates an undesired physical intimacy with strangers. Increased interactions between strangers in business transactions acts similarly, though the contact is not as extensive. Overall modernization and urbanization bring a decrease in control over with whom one comes into contact through an increase in the number of people with whom one interacts and an increase in the anonymity of individuals.

What, then, is caste today? I can answer this question only in terms of what I observed. I approach caste as a practice, rather than a model, which is defined through everyday relationships and interactions. And for Nepalis, jaat may be thought of as some

or all of the following. Jaat is a frame of reference used and twisted to fit one’s

purposes, a fallback when uncertain about the rules governing social interactions, a

means of networking and making claims on other families of the same group, an identity

to be exploited politically. It is an expression of deference by the powerless, such as

rinsing out one’s own cup after drinking tea in a public tea stall instead of expecting the owner to do it, or waiting until last to take one’s share of drinking water amongst a mixed-caste group of agricultural workers. For low-caste individuals, it is an outdated

source of stigma that is offensive and threatening, and a potential source of belittlement

to those who are successful or in positions of power. It is an oppressive system that has

been delegitimized publicly, but that still has power to influence people’s behavior.

Rethinking Women’s Autonomy and Caste

Feminist scholars have underscored the diversity of women’s experiences within a single cultural context, and the danger of assuming that the dominant model of what it means to be a woman applies to disadvantaged groups. Gender norms are defined differently according to jaat. A greater concern with ritual purity and honor among high castes leads to more restrictions being placed upon women. Although married couples in general are particularly prone to polluting behaviors as they must reproduce and are often involved in agriculture or artisan production during the householder stage of life

(Cameron 1998), much of the ritual purity of a household falls on the shoulders of its

women. Women have a greater risk of polluting a household because of their

menstruation and the bodily substances associated with birth. Some groups consider

women’s bodies to be more open/less bounded in general, and thus a better conduit for

77 pollution (Lamb 2000). Given that this is the case, it follows logically that high castes would need more rules to regulate women’s bodies and their behavior in order to maintain a ritually pure and respectable household. Thus caste has had an important determining effect on women’s roles, with restrictions on women’s autonomy increasing as one moves up each level of the caste hierarchy.

In her study of gender and caste in far western Nepal in the late 1980s, Mary

Cameron investigated the lives of low-caste women in order to articulate the ways their lives differed from the dominant discourse on gender and high-caste women. She demonstrated that although low-caste women are ritually impure according to the high- caste perspective, a different set of ideals supersede the purity/pollution ideology of caste for low-caste women. Their contribution to subsistence and wage-earning activities allowed them a social autonomy and relative economic power that most high-caste women did not have. The economic necessity for women to be engaged in labor related to agricultural work or other small-scale earning activities ultimately freed them from some of the patriarchal restrictions on mobility and interactions outside of the home.

Although Cameron argued for a more nuanced reading of low-caste women’s autonomy than the simple claim that low-caste women have more freedom and autonomy than women of high caste, she concluded that low-caste women’s income and the social benefits of their labor give them more leverage than upper-caste women who depend on honor for their ideological power (Cameron 1998).

Women in Vishnumati disagreed over the implications of caste for contemporary

Nepali women. Some women said that it depended on the individual, their education,

their economic status, or whether they had adopted an urban lifestyle. “Well it depends

on the actual life conditions of the people, how can we say (how life differs for low- and

high-caste women)?” Others were hesitant to answer, but made general statements like,

“People say that the life of lower caste women is freer than the life of higher caste

women. For example the lower caste can do any job and go anywhere.” Some high-caste women expressed the opinion that low-caste members do not have to be as respectful in the forms of speech that they use, using the “timi” form of address. Both the generalizations and the caveats that women articulated foreshadowed the conclusions that

I would reach after analyzing the case studies along the lines of class.

The experiences of low-caste, low-class women in Vishnumati bear out

Cameron’s findings on low-caste women’s rejection of high-caste gender rules.6 In this sub-group of families, women did not have to show as much deference to their husbands through respectful forms of prescribed speech and behavior. All of these women were

engaged in wage-earning labor. These families were concentrated in the bazaar area –

they were renters, not landowners, and were not involved in the agricultural sector like

many of the high-caste land-owning families. Three women worked as tailors in self- or

family-owned businesses and operated one-room shops. The women managed all aspects

of the businesses: taking orders, sewing, and negotiating prices. One very poor woman

managed her own fruit shop, selling whatever was in season at her roadside stall. She

also handled all management aspects of the small business, often traveling an hour’s

distance to the market where she purchased her fruit and setting and negotiating prices

6 For the reasons explained previously, all of the low-caste families included in the case studies were of low class.

according to various market influences. These women were small-scale entrepreneurs

acting mostly independently of male involvement.

The other low-caste, low-class woman in my case studies worked as a janitor in a

hospital. The nature of her work was much less entrepreneurial than the former

examples, but it resulted in similar exposure to people beyond her community. She had

to travel by bus to reach her place of employment, and she received a set amount of

wages for her work. This woman was often present in the bazaar – I ran into her in

public spaces frequently, whereas with all the other women I saw them on the streets only

once or not at all. She was also the victim of gossip among locals because she used to

work the night shift cleaning a hospital, and had to return home late in the night.

Noticing this, people gossiped about the possibility of her being a prostitute. Her

freedom of movement, especially at night, had consequences in the eyes of the public.

This last example draws attention to some of the nuances and complexities of the

interaction of gender and caste for low-caste women. Viewing gender norms for women

from a low-caste perspective is necessarily taking a subaltern perspective on caste. From

the subaltern perspective, low-caste women do have more autonomy in matters such as

getting a job as a janitor and independently traveling to and from that post. But in the

views and reactions of the larger community, from the dominant perspective, this

woman’s transgression of the dominant (read high-caste) perspective on gender roles

results in a backlash of the public/members of the community speaking against her moral

character. It is clear from these public expressions of disapproval that, from the dominant perspective, this woman should not be working in this fashion. Thus to say that low-

caste women are exempt from the gender rules of the dominant/high castes is to miss this

80 matter of perspective and the disapproval that is expressed from the larger community or society. One low-caste woman said,

Society wants a woman to stay inside the house and not speak with anyone, otherwise they start backbiting you. Even if we wear nice clothing they will talk about us and start commenting that so-and-so’s daughter-in-law is behaving like this and that. We can’t speak with outsiders…

Such a statement sounds like it should have been spoken by a high-caste woman. Going out, wearing (and thus flaunting) nice clothing, and speaking with non-family members are all acts of immodesty and may bring social scorn upon a woman. The distinction between the experience of gender for low-caste women, then, is their (partial?) rejection of dominant society’s gender roles while, at the same time, they are looked down upon for doing so by the larger society. They operate in dual systems of gender, for they cannot escape the dominant society’s perspective. While acting within one system of gender, they face repercussions from another.

In line with Cameron’s findings, among my case studies it was high-caste women who were restricted the most from leaving the home, speaking excessively with family members or gossiping with neighbors, appearing “over-smart” in dress or speech, and touching certain religious or food items while menstruating.7 One soft-spoken, young

Chhetri woman confided that she felt a little confined within the home of her husband’s large joint family. The family did not like for her to go outside, so she rarely left the home even on errands and never left to just walk around (gumna jaane). Another woman around the same age, a Brahman woman, explained that women should behave in this way:

7 More true for extended family households than nuclear.

Women have to be a little bit down. They don't like women to speak like they are important. Women have to mind others and elders. We have to obey the husband. We have to endure, to mind whatever is said, no?

High-caste women were expected to express the most deference toward their husbands

and husbands’ families.

But I also found contradictions to these generalizations about high-caste women.

Through comparing poor and middle class families within each caste, it became clear that

poor high-caste women were not bound by the same rules. Out of economic necessity

some high-caste women of lower socio-economic status worked as laborers in the fields

or assisted with large-scale food preparation for special occasions for other wealthier

locals. These high-caste women who regularly engaged in field work were the most

difficult women to meet for interviews. They always seemed to be out planting rice in this neighbor’s field, or hoeing a plot of land for that person’s corn. Thus being confined to the home was hardly a problem for this group of high-caste women – they were rarely at home, especially during peak agricultural seasons. These were also the women that could be found in local tea stalls during breaks, a public space usually only occupied by men.

The significance of class for high-caste women can be observed within the walls of a single building in Vishnumati. The large, rectangular, three-story building is owned by the family of the Chhetri daughter-in-law mentioned above. Four generations of her husband’s family live in the building on the second and third floors. Most of the first floor is rented to several other families. In one corner of the first floor is an informal tea stall/restaurant where locals often gathered for a snack during breaks in the agricultural workday. The Chhetri daughter-in-law rarely left the top stories of the building, spending

most of her time involved in food preparation for the family. A Brahman woman, just a

few years her senior, lived downstairs with her husband and their two boys in two rooms

that they rented on the ground floor. In contrast, this woman was rarely at home. She

worked as an agricultural laborer in order to earn money to help pay for clothes and

school fees for her two boys. Her husband was a teacher, but his income was not enough to support even their meager existence. The only times she was at home were at dawn and again in the evenings after dark. She had taught her young sons how to cook because

of her regular absence.

The economic necessity for her to work outside of the home set in motion a series

of consequences that disrupted gender roles and restrictions for high-caste women. First,

allowing her boys to shoulder some of the responsibility of food preparation signifies the denial of one of women’s most symbolic and time-consuming responsibilities, cooking for the family. Of further interest, the sex composition of her children demanded that sons, not daughters, take on that responsibility. Second, her work for various local land-

holding families required that she interact with men and women from the community and

those nearby. She also occasionally traveled to nearby towns when news spread that they

needed extra hands. Although she may have taken care to maintain a sense of modesty

throughout such interactions, her mobility and public presence threatened the traditional

gender expectations so readily visible in the behavior of the Chhetri family upstairs.

This example begs the question of whether the poor Brahman woman has greater autonomy than her Chhetri neighbor, and raises a much larger issue of the extent to which autonomy and other aspects of women’s well-being may be in conflict. The Brahman woman had few restrictions on her movements outside the home and was a valued source

of income for her family, but she had little leisure time, engaged in demanding physical

labor, and was separated from her family. Women who move about alone in public may

face indignities while working outside because they are less protected, easier targets for

harassment (because of the lack of respect for their autonomy). The Chhetri woman had

less autonomy, but she had a more comfortable life in terms of her basic needs, such as

adequate food and shelter, being met. She had the luxury of only doing light housework

and having more leisure time, but she rarely left home. It was clear that she longed for

more social contact.

In this social context, poverty acts as a leveling mechanism between high- and low-caste women, making it impossible for either to live by the dominant set of high- caste gender ideals. What are the implications of arguing that poor high-caste women and low-caste women have more autonomy, while as a result of their poverty, they engage in physically demanding labor or have difficulty providing and consuming nutritious foods? To consider autonomy as the primary indicator of women’s well-being would be to miss the effects of poverty on women’s overall health and fulfillment. When talking about women’s autonomy, other measures of discrepancies in power and social hierarchies must also be invoked, and attention paid to how they interact.

Like autonomy, education did not appear to be a good predictor of women’s well-

being or quality of life. Women who were educated through the secondary level led

similar lives to their less-educated counterparts. After marriage, women who had

attended secondary school and who could have sought wage-paying jobs in the informal sector did not pursue such work outside of the home. Sita, a young Brahman woman who

married into a middle-class extended family, had studied through class nine. In the first

few years of her marriage, her husband’s family owned a sizeable amount of agricultural land, and she spent much of her time helping the family with agricultural tasks. She said,

“Even though women are educated, society says that [women should stay down in

Nepalese society]. Society says, ‘What will they (women) do if they are educated?!’

People think that after being educated that they will accomplish something, but ‘What

will she do?!’ is said by society.” As Sita’s words attest, being educated did not

necessarily result in a different set of work options or non-domestic roles for young

married women in their 20s. Sita’s story will be told in more detail in Chapter 5.

Durga, also a young high-caste woman with a secondary level education and a

middle-class joint family, lived in a large house with a private courtyard along with her

many family members. Altogether there were nine family members, including two

unmarried daughters and one unmarried son. Thus she was the only daughter-in-law

(buhaari) in the household. This meant that most of the housework fell on her shoulders;

she joked that she spent her entire day in the kitchen. She told stories of being forced to

cook for the family despite intermittent leg and back pain. She said that once, “I was

shaking, but even then I had to work.”

Durga relished telling that she had attended school up to the SLC (School Leaving

Certificate) level, while her husband only studied through class nine. She was married

young, by her own description, as the eldest and only daughter in her family. During the

arrangements for her marriage, her father had demanded that she be allowed to attend

college after marriage. Her parents-in-law disagreed, but ultimately she ended up failing

the SLC exam and thus did not have the option to continue on to college.

After several interviews, during a discussion of how strict her parents-in-law

were, she casually mentioned that she once drank an entire bottle of medicine in order to

poison herself. Her anguish had reached the point where she protested through

attempting to harm herself. She said she became so frustrated that she did not want to live anymore. Later in the interview after a follow-up question about the incident, she

explained:

My mother-in-law used to shout at me on one hand, and on the other hand my husband did not love/care for me. What will happen to my mind? Because of that I drank medicine (oushadi). She was always shouting at me and my husband was not looking after me. After I got my daughter, I did not care about them, whether they loved me or not. I just loved my daughter.

She relayed this story in her normal way of speaking, with a subdued liveliness and

congeniality. She admitted that she did not think she would actually die from drinking

the cough medicine, but that she felt like she wanted to die. She felt trapped and unhappy

because of her family situation, and her education and experiences prior to marriage

made her more aware of her situation and discontent. For one of the first generations of

women in which a secondary education was becoming standard, being educated seemed

like somewhat of a curse. It changed women’s ideas but not their circumstances.

These examples illustrate that although at the aggregate level caste is correlated

with opportunities in life such as education and wealth, at the individual level there are

exceptions. Micro-level analysis reveals that among the higher castes as well as the low,

women’s capacity to earn is recognized and utilized by families of lower socioeconomic

status. As a result, the control of women’s movements and communication with others is

loosened in these families. I argue that low-caste women’s autonomy should be

understood in the context of their economic status – historically low – and not their ritual

status. The pragmatics of survival among impoverished people preclude adherence to

strict gender norms that limit female participation in productive work. The similarities

between the relaxation in/transgression of standards for/by poor high-caste women and low-caste women lead me to this conclusion.

Future research needs to address the situation of middle-class and wealthy Dalit women. Do they adopt the gender standards of high castes and reject the ways of poor, low castes? Or does their autonomy, combined with their high economic status, allow them to surpass high-caste elite women in education, careers, and income? As I mentioned in the methodology, I had difficulty finding low-caste families who were middle class – statistically there were few, and the two families that qualified chose not to participate in the case studies. Could these two families be examples of attempts to advance socially by adopting similar behaviors and trappings of the traditional elite? It

seemed as if they did not want any attention drawn to them by participating in the study,

potentially out of a concern to fit in with high-caste families and to avoid distinctions

between them and their economic peers.

Contemporary young Nepali women are educated, especially in more urban areas,

and they have different views from the previous generation about women’s roles and

rights. But in extended families, especially high-caste, middle-class families in this semi-

urban community, young mothers must defer to their elders and their husbands in matters

of work, fertility, and birth. As class becomes more influential in people’s lives than

caste, the effects of education and socioeconomic status will start to look more like the

pattern with which social scientists are familiar: a positive correlation between women’s

87 education, economic status, and autonomy. But it will take a generation for young mothers’ new ideas about gender roles to manifest in changed behavior because of the restrictions placed on their autonomy by the elder generation.

CHAPTER 3

FROM DAUGHTERS TO MOTHERS-IN-LAW: WOMEN’S POSITIONALITY OVER THE LIFE COURSE

In addition to caste and class, a woman’s stage of life significantly impacts her position and the definitions of gender for her during that stage. This chapter transitions from the previous discussion of social hierarchies to a look at familial hierarchies and the life course. The observational portion of my research was conducted with a synchronic perspective on family hierarchies: I observed the relations among family members within a delineated period of time. Thus my observations constituted a slice of family life, with different members located in different positions. My retrospective interviews with women, however, captured a different perspective: they follow women in their progression through the relevant stages of the of the life course, which, in this particular context, were largely defined in terms of family life. The domestic cycle and the life course stages of women were intertwined. When employing a retrospective account in this way, I take into consideration the historical context and changes over time.

The dominant family system amongst Hindu Nepalis, along with much of Hindu

South Asia, is the multi-generational extended family, with patrilocal marriage practices and patrilineal inheritance and kinship. This does not mean that all families are in fact joint families at any one point in time. Fricke’s research for example, conducted with a

Tamang group not far from my research site, exhibits in impressive detail the cycle of

89 joint families and how married sons split off from the stem family when the number of family members becomes too great for maintaining harmony and sharing resources as a single unit (Fricke 1994). Thus a number of families will always be in a nuclear stage within a joint family system (see Skinner 1997), and detecting a trend towards a conjugal family system is tricky. It would take another generation or two, for example, to determine if the nuclear families in my study would remain nuclear. More important for this study, in the end, was their belief that they would become joint families again.

Demographic outcomes such as fertility and health are significantly impacted by kinship and family systems. Das Gupta’s apt description of the north Indian joint family system applies historically to the Hindu groups with which I was working in Nepal as well:

…there are strong intergenerational and intragenerational bonds between household members related to each other by blood. Concomitantly, the development of a strong conjugal bond is discouraged. This means that the woman marrying into the household is in a very weak position in terms of making decisions to protect her own and her children’s health. Layers of people are above her in the household hierarchy of status and authority, beginning with all the adult males, and continuing through all the women older than her (1997:43).

As mentioned earlier, the site of this research was selected for its location amidst a transition from an agricultural to a cash based economy. The inhabitants of Vishnumati were undergoing a shift from a familial mode of production to a capitalist mode of production, even though actual opportunities for income had not caught up with the increase in the value of education. Slight challenges to the hierarchy described by Das

Gupta were visible as a result, such as the undermining nature of a higher level of education of daughters-in-law in comparison to that of the mother-in-law in a household.

The cultural construction of being a girl and a woman are also highly influenced by the dominant kinship and family systems. Among Hindu-caste groups in Nepal, and particularly among high-caste groups, symbolically women are not fully members either in their natal or their marital homes (Bennett 1982). Being a daughter is like being a guest in a household that provides care and nourishment, for a daughter will inevitably leave the household and her consanguinal kin. As a child, a daughter is cherished as an embodiment of the goddess, as a pure being. Although the protection of her sexual purity is a major duty of her parents, a daughter is fairly free within the confines of her home to do things like speak her mind and to wear comfortable and casual clothing. She is not responsible for the future of her natal family through progeny, name, wealth or success.

Though the laws have changed to allow women to inherit land, Nepali Hindu society remains overwhelmingly patrilineal: inheritance and name are passed down through men.

As a wife, a woman is symbolically responsible for producing the descendants of her husband’s family and protecting its good name through her piety and devotion. Her claim to the family name, though, remains an indirect one: it operates through her husband. In the case of early widowhood or separation, a woman’s affines may easily shirk even the most basic culturally deemed responsibilities toward the woman such as shelter and food. Even women’s inheritance from consanguinal kin given to her at the time of marriage (daitwa) is often forfeited should a woman ever seek the rare option of leaving or divorcing a man (Weiss 1999). Thus one could argue that a woman is never truly at home in either in the home of her father or her husband, or in the case of old age, of her son.

Life as a daughter, labor as a daughter-in-law, uncertainty as a mother-in-law

For women, marriage represents a physical separation from people and place, but also a symbolic shift in points of reference in the most intimate of ways: a redefining of self, family, and home. That such changes come with marriage is not unique to Nepal, but the ways these transformations take place are culture and context specific and consequentially have particular outcomes in terms of women’s autonomy, health, and fertility.

In order to portray both the commonalities and the potential for individual uniqueness of women’s experience of the transition from daughter to daughter-in-law, I use one woman’s detailed story at the outset of the sections on marriage, learning to be a daughter-in-law, and becoming a mother-in-law. Much of this chapter centers on Maya’s life history as a way to evoke women’s life course experiences and create a structure for examining the most relevant points common to most of the women in study. When I interviewed her in 2003-2004, Maya was 37 years old and had two daughters and one son. Her eldest daughter was studying in college and working part time as a teacher. Her husband had a salaried job, and the family was middle-class and of the Chettri jaat. They had few outward markers of the middle class; they lived in a modest house on a small plot of land, and they still owned some agricultural land some distance away. In the segments of her story, I try to preserve the language that Maya used in telling it, and I occasionally break into the narrative with her exact words for emphasis or when an

English equivalent does not do justice to the meaning of the original Nepali words or expressions.

A daughter becomes a bride

Maya first met her husband on her wedding day. At first her mother had said that

she did not want to give her daughter, Maya, to anyone who lived in BNK. The place

was not “developed” (bikas) at all at that time – merely a small village. But then after seeing her future son-in-law, she changed her mind. She liked him.

From the prospective groom’s side and from Maya’s side, representatives of the

families met at the groom’s house. In the beginning her mother had said that Maya

should not be given away to such a place, but her relatives recommended that she go to

see what it was like. Reluctantly her mother went to the boy’s house, saying that she

would go only as a formality. But after seeing the boy, on the way back home, she felt

that he would “raise” (paalcha) her daughter well.

Maya’s prospective husband had six brothers; despite that being the case, her

mother decided to give her to one of them. (This meant that their property would be

divided among the 7 brothers, and each one’s piece would be small.) Maya’s mother

began planning to give her away in marriage, and she informed Maya. Soon phupaju

(Maya’s uncle) and bhinaaju (her brother-in-law) went, gave tika, and left. “It’s like this

for us, our tradition – give tika to the boy, fix the details of the wedding, give janai

(sacred thread) and supaari (betel nut), and ‘From today our daughter we have given to

you’ has to be said by the mother to the future husband.’”

At that time (around 1983), people did not meet before the wedding. Some

people probably did, but she and her two sisters definitely did not. The wedding

happened “all at once” (ekai chutti). No one from the husband’s side came to Maya’s

house to arrange the marriage, but Maya’s was an unusual case. She had no brothers,

93 only sisters, and her father had passed away while she was still in her mother’s womb.

Seven months after his death, she was born. Her father had spent all the inheritance before passing away, and Maya and her mother stayed there in the village until she got married. After she left her village, Maya’s mother lived with Maya’s older sister’s family

(for her mother had no sons).

For the wedding they called all of the villagers in Maya’s village. A bus load of guests had arrived from the groom’s side, plus around thirty local wedding guests. There were only fifteen or sixteen houses in her village at that time. She had not fixed her makeup or her hair when the janti (groom’s procession) arrived. She was sitting outside just like that. And then someone said, “Well, the groom has already been brought and you haven’t done your makeup.” Then friends came to arrange her hair for her, and her mother told her to put on everything. And her mother asked if she needed anything to comb her hair. They put on the makeup, the wedding occurred, and in the evening they took her. She returned with the guests, musicians, and all in that bus. They reached there around eight o’clock at night, and had to walk after the road ended. She had had only a glass or two of water. They said to feed her, but she did not feel like eating. It was a different kind of food than she was accustomed to at her old home, and she wondered how she could survive eating such food. It was very uncomfortable for her. Leaving her own house was difficult. One does not know what it will be like. She was seventeen when she got married; it was not so early. But she cried.

“Why wouldn’t one cry! Leaving one’s own home and going to another, won’t a person cry then?!” Before the wedding there had been only the two of them living together, mother and daughter. Whenever Maya went out, her mother used to wait for

her to return until she made tea. Her mother always used to wait for her before eating

anything. After giving tika to Maya’s future husband, her mother did not cry in front of

her, but behind her back she cried loudly – every day after fixing the date of the wedding.

Thinking that Maya would hear, her mother would go to the garden to cry. In one place

Maya would cry, and in another place her mother would cry, separately.

…

Daughters in Nepali Hindu families are shielded by their mothers from the harsher

aspects of life as a woman, including work. Daughters have responsibilities in the

household from an early age, sometimes from age six or seven, but children are mostly

indulged before that. Young daughters learn to wash the metal plates and cups, wash

clothes, and shadow their mothers in the kitchen. For the most part they are not asked to

do physically demanding work. This is contingent upon the socio-economic standing of

the family; in very poor families children must shoulder, often literally, much physical

work. Poor young boys are even employed in the informal economy outside the home in

positions such as fare collectors on public transportation.

Daughters help out in household work, and they are chided if they do not or if

they act “lazy;” but as Maya said in the next part of her story, daughters have the option

of saying “I can’t.” This important distinction applies to a range of situations, including

becoming tired from work, not knowing how to do a particular job, or having a

competing obligation. I observed numerous situations in which adolescent daughters

who had an important test coming up in school were able to skip their normal household duties. Alternatively, if agricultural work such as carrying heavy sacks of rice or planting is unappealing or seems too difficult, girls can decline. This is increasingly true with the

95 changes in ideals among the middle class. Involvement by young people in agricultural labor is increasingly looked down upon by those who have the means to avoid it.

In the end, girls are not responsible for making sure the work gets done and the household runs smoothly – that responsibility falls solely on the mother (who may be a daughter-in-law in a joint family household). In fact, the differences between the status of a daughter and a daughter-in-law are most palpable in the differential standing of the two women in a single multi-generational family. In a situation where there are unmarried daughters still living at home with a son, his wife, and his parents, the daughters in the family have very few responsibilities or concerns because they are all handed over to the daughter-in-law. In several situations of this kind, the daughter-in-law seemed more like a hired domestic worker in the home than a family member. The daughter-in-law adopted behavior like not speaking and looking at the ground when in the presence of a daughter of the family. And countless times I overhead daughters yelling for the daughter-in-law of the house, calling her by her kin name “Bhauju!” and yelling orders that permeated the rooms of houses, about bringing tea, fixing snacks, and locating this or that missing object for them.

Maya’s description of her mother waiting for Maya to drink tea and eat with her is indicative of the special treatment women receive while their main identity is that of a daughter. Food and eating are highly symbolic in Hindu cultures, and Maya’s inclusion of this detail in the story of her life as a daughter would be understood as quite significant to another Nepali or someone familiar with Hindu cultural symbolism. The conveyance of pollution (juTho) through passing or sharing food or water from an individual with polluted status to a ritually pure person as reviewed in Chapter 2 is only one of the ways

that food is central in familial and social relationships. In addition to the purity and

pollution rules governing food and especially cooked rice that remain widely

acknowledged in South Asia, the act of eating and the order in which people eat is of

great significance. Although not always followed by the families with which I was the closest and in which I had the opportunity to observe occasional exceptions, women of lowest status in the household eat last after serving the other members. For example, the

daughter-in-law of a joint family would not eat until the other senior members had been

served or had finished eating, and the mother in a nuclear family would do the same. I

would be remiss to reduce this practice to merely the oppression of women and the denial of choice foods and nutrition to women – although there are real aspects of this operating and a potential for harmful consequences for pregnant women. A nuanced explanation of

this behavior pays attention to women’s emotional and strategic reasons for eating last.

This is one of a range of behaviors that women use to demonstrate their devotion (and

love?) for their husbands and families. This behavior may be felt genuinely and deeply

as an expression of devotion, or it may be resented as merely a duty depending on the

specifics of a woman’s relationships. On the whole, feeding someone (meaning serving

someone food) is a poignant expression of caring for them. A distinctive example of the

significance of feeding someone is found in the actual placing of food in another person’s

mouth by hand, in the same way a parent feeds a young child. This occurs during a

marriage ritual in which bride and groom publicly overcome the juTho rules and feed each other using their right hands. Given the cultural importance placed on food and eating, it is not surprising that not eating a meal is a dramatic expression to the immediate family members of hurt, anger, or any kind of severe distress.

Rather than a day of celebration and the symbolic joining of two families, the

wedding day in Hindu cultures in South Asia has traditionally been a day of solemnity,

separation, and the transition to a new family for women. A daughter is given to a new

family through the ritual of marriage (Fruzzetti 1982). On the day of marriage, first the

groom and his family and friends travel in a lively procession (janti) to the house of the bride. The bride’s family hosts the wedding rituals and reception dinner. The various stages of the wedding call on the bride and groom to sit somewhat passively while gifts are presented, their feet are washed, and a priest recites scriptures. While this is happening, brides characteristically sit with downcast eyes and a solemn, if not sad, expression. Rather than merely following a social script on correct behavior, women explained that they were genuinely sad on the day of their wedding. One woman explained, “You feel like crying, and so you have to bow your head… it’s not a tradition, you really feel like crying.” Her Budhi Saasu (her husband’s father’s mother) added, “Of course you feel like crying – upon leaving your mother and father, you will feel like crying.” Almost all of the women in the case studies reported crying during their

weddings. Some older women said they did not cry because they were too young at the

time of their wedding to understand what was happening. Several women compared

getting married at a young age to playing with dolls, getting a chance to play dress-up in

fancy clothes. However, being taken away from home was different. “I felt as though

we were playing dolls. I was still small. Even after I was brought here [husband’s

house] I kept telling them that I wanted to go to my own home. I cried when being sent off. I felt, where I was being sent away?”

Added to the separation from family and friends, and general familiarity with

things, is the fact that most young women in my case studies had never spoken to the

groom before their wedding day. They may have seen the groom when he came to “see

the face” of the bride and finalize the marriage agreement, but never were words

exchanged. One woman exclaimed, “How could I speak to a person whom I didn’t

know?! It was not like now; things were different. We used to run away if some stranger

tried to talk to us. We were shy. Only now people talk first and get married.” Themes of

separation from family and being entrusted to strangers filled women’s stories about marriage.

The experience of separation described above was typical for the women in my case studies who had arranged marriages. Even the couples who split off from the stem family within the first year of marriage lived at least initially in the joint family, so that women had somewhat similar experiences of departure. The other few women who had

“love marriages” experienced a different kind of separation – some were escaping an unhappy relationship with family members and therefore glad to leave, and some were outcaste by their families after the knowledge of their relationship with their husband became public.1

Marriage represents a separation of almost everything that is familiar to a girl –

the love and protection of her parents, the comfortable spaces that exist within the walls

of the home in which she was free from many of the public gender restrictions on

behavior and appearance, and her relationships with family members, neighbors, friends,

and places in her hometown. The separation and anxiety that women experience at the

1 It will be necessary to watch how an increase in inter-caste and other “mixed marriages” will alter the description of marriage presented here.

time of their weddings are parts of a rite of passage into womanhood. Another part of the

transition to marriage is a dramatic change in social and kinship roles. Almost overnight

women gain all kinds of new responsibilities, many of which are unfamiliar. For an

explication of this aspect of learning to be a daughter-in-law, I return to Maya’s story.

Learning to be a daughter-in-law

After Maya came to her new house, there were a lot of problems. Even though

Maya’s mother had encountered difficulty in life, she had made sure that the children

were happy. At her maiti (a woman’s natal home), Maya did not have to do that much

work. At her new house there were many family members, a lot of livestock, and many

fields. Maya was not accustomed to such work, especially carrying a heavy load for a

long distance. She began her duties within her new home on the third day after her

marriage. She went to the potato fields to work, and she had to carry a dhoko of compost.

“I was not able to carry it, and two others helped me carry it. Still they laugh at me by

saying that three people helped to carry it… I couldn’t even walk properly from here to

there in this steep area while wearing the naamlo.” It was very difficult for her to start

such work suddenly as an adult.

There is much difference between maiti and ghar.2 There are differences in work, food, dress, and going out. “In maiti we don’t have to work if we can’t. We can say to

Mother that we can’t do it, and that is alright. ‘I don’t know how’ or ‘I won’t’ can be said. But here [in the husband’s home] that doesn’t work. We have to do it even if we don’t know how or we are not able [physically] to do it. That is the difference.”

Everyone else in Maya’s ghar could do more work than her. And that did not work, of

2 Here ghar means a woman’s husband’s home. A detailed explanation of this term will follow in the text.

100

course, in the household. It took a long time for her to wash clothes, and in that amount

of time others could finish a lot of work. The others in the family did not wash clothes

very often. No matter how dirty the clothes became, they kept wearing the same dirty,

torn clothes. Maya’s different standards for clean clothes became a major point of

contention in the family.

There is also a vast difference in the behavior of saasu and mother, of course.

When Maya’s mother scolded her, she did not feel bad. When her husband’s relatives

scolded her, she felt unhappy. It was different with her own relatives.

…

Marriage marks the end of being a daughter in many ways, although women

return to their natal homes whenever they get the chance. The natal home remains a safe

haven for married women especially during socially sanctioned periods of visitation such

as Dasain or after giving birth. After getting married, a daughter’s responsibilities to her

natal home are severed for the most part, other than the symbolic responsibility of the

Bhai Tika ritual. Women travel to their natal home for this ritual during Dasain. A sister offers her brother(s) blessings through a ritual plate of special food items and giving him tika.

With marriage a woman’s allegiance and service switches over to her new family, the patriline. A woman’s main identity thus changes from daughter (chhori) to daughter-

in-law (buhaari). This is reflected in how others address a woman: when calling her,

younger people in the community commonly address a married woman as “Bhauju,” meaning elder brother’s wife. Kinship terms supersede proper names.

101

A woman’s main identity is not the only name that changes. That which she has called her “ghar” (home) her entire life she refers to as her “maiti” (natal home/a

woman’s father’s home) after marriage. The word “ghar” for a married woman refers to

her husband’s house and his family members that live there. Furthermore, “ghar” has the

connotations of the English word “home” in the sense of being from or of a place.

When a bride first arrives at her new home, she is often met with special treatment

for at least the first day. After that, she is introduced to her new responsibilities or

sometimes left on her own, guessing and worrying about what she should be doing in the

new house. Women talked about a constant uneasiness in their new ghar, which

stemmed mostly from the strangeness of the new place and people and worries that they

would not do tasks the “right” way.

Becoming a wife and a daughter-in-law brings major changes in terms of a

woman’s responsibilities, both in type and in rigor. Many young women had never been

involved in agricultural work before, and it was especially foreign to those girls coming

to Vishnumati from the urban neighborhoods of Kathmandu. In the late 1980s and 1990s

when most of the women in my case studies were getting married, Vishnumati was rural

and “undeveloped,” as described by Maya’s mother. When I first visited Vishnumati in

2000, there was a dirt road leading up the mountainside to the village and only a small

bazaar area near the temple. Most of the homes had much wide open space and fields

between them. When I returned in the spring of 2003, I hardly recognized the area that I

had been only three years earlier because of all the new buildings that had been

constructed and the paved road. And within the 11 month span of my longest

102

consecutive research stint, between the fall of 2003 and the summer of 2004, houses were

constructed at a visibly fast rate.

Back when many of the women first moved to Vishnumati it was little more than

a forest, according to one Brahmin woman from Kathmandu. When she first arrived in

the village after her wedding, she cried and cried upon seeing the surrounding forest.

“There were such big boulders everywhere,” she said, “but nowadays nice houses have

been built all around, and it’s nice now.” Gathering firewood in that forest for cooking,

undertaking agricultural work, and raising livestock were daunting to the women from urban families. Another woman raised in Kathmandu described the early years of her marriage, during which she had to gather firewood and fodder to feed the animals, saying, “I rarely talk about this with anyone. About the past I want to cry, it was so sad.

All over (gestured toward hands and arms) there were wounds. I had to go to the forest, and my feet and legs were scratched all over. It used to be like that.” Women in farming families work hard in the fields; only plowing the earth is restricted to men. Married

women of low economic status are forced into positions of paid hard labor such as carrying loads of construction materials or fodder, or doing the most back-breaking parts

of farming for land-owning families.

The two women quoted above stressed how much more pleasant their lives had

become after they were able to quit doing work like transplanting rice and gathering

fodder from the forest. These changes and overall improvement in the women’s well-

being were unrelated to life course effects. Historical developments, such as changes in

the economy and the growth of the education system, encouraged a general move away

from agriculture as the preferred source of livelihood. Individual-level improvement in

103

the economic status of some families, while related to opportunities created by an expanding economy, meant that an income from a wage or salary-paying job could replace toil in the fields.

Improvement in Maya’s position was related to her husband getting a modest salaried job and leaving the joint family; thus she was able to leave behind arduous agricultural tasks and her share of the duties in a large joint family. She attributed their separation from the joint family household mostly to her inability to work, but her husband’s salary actually enabled them to start building a new house. In her case it was a combination of economic changes and a change in household structure that precipitated the change in duties and status.

Becoming a mother-in-law

Maya could not imagine what it would be like once she became a mother-in-law and brought a buhaari for her son (who was 14 at that time). She laughed and said that she would probably have many problems after bringing a buhaari. She figured that she was not able to make her Saasu happy in the past, and so it might be the same. The buhaari who comes might satisfy her even less than she satisfied her Saasu. That’s a possibility. She had not dare to hope that her son would grow up and bring a buhaari through marriage and give her comfort and happiness. She smiled, and modestly admitted that she might be the same as her Saasu after she became a saasu herself. “What will happen, how can we know? As much as possible I have to think it will not be like that, no? But tomorrow, anything can happen in the relations of saasu and buhaari. It is impossible to keep good relations. There is little possibility of good relations.”

104

Maya did not think that after she got a buhaari that she would be able to rest and have an easy life. She thought that her current life was easier, and wondered what would

happen in the future. Society says that one needs a son and daughter, but that it is just a

saying and she does not think like that. However, she already had a son, and she thought

that her thinking might be a result of that fact. “If I had not had a son already, I would

have also felt like that maybe.” Maya thought that it would be best if her son and buhaari could live together (with her and her husband in a joint family) and keep a good relationship. But if they could not, then they should separate. But it would be best to stay together if they could. She intended to treat her buhaari well.

…

Near the conclusion of many months of interviews with Maya and the other women in my case studies, I asked them about their hopes for a future daughter-in-law.

Like most of the women in my case studies, Maya hoped that after her son married he would continue living with her, effectively turning her nuclear family back into a joint family household. She was ambivalent about how this would turn out, however, pointing to her own highly contentious relationship with her mother-in-law. With a keen sense of self-awareness, she joked that she might not seem any different to her future daughter-in- law than her mother-in-law had seemed to her.

Maya’s improvement in her quality of life was related to the split from her parents-in-law’s joint family household and from a life of agricultural labor.

Alternatively, women who were still involved in heavy agricultural labor traced the improvement in their positions to change in their status within the household rather than

escaping agricultural work. They attributed improvements in their quality of life to

105 having capable children and to becoming a mother-in-law. But like Maya, women living in joint families and doing agricultural work were of two minds about modern daughters- in-law.

A Brahman woman, the wife of the eldest son in a large joint family and the mother of four, exemplified the life of a woman engaged in demanding agricultural production. Shrimati’s husband’s family owned the most agricultural land out of all the case studies, and her condition represented one extreme of a range of the degree of physical labor performed by women. Her husband had a high-profile job in Kathmandu but made little money, so she, the other daughter-in-law in the household, and her children were responsible for almost all of the work in the fields and for the livestock.

The children were attending school, and the younger daughter-in-law was pregnant; most of the physically demanding work thus fell upon her.

Shrimati had strong opinions about the ways economic developments affected men versus women. When asked about her duties in the household, she responded:

What can be done. There are fields. Even if I had studied, then what?! There are lots of fields to look after. Daughters are made to do farming work in the fields, and now look at the sons – they are lawyers, business men and so on; and look at the plight of daughters – they are married and sent off to the husbands’ houses so if the husbands give them troubles they what will happen to them. If women had studied and become independent then they would be able to stand on their feet and not be dependent on their husbands. Now if the husband ill treats them or beats them they are helpless; they have to stick with such men. What I mean is, women have been disadvantaged. Now the men own vehicles, motorbikes and lots of lands. They are dominant personalities in the village. But even then, though they had all that, they did not educate us girls at all. Why? Because they had so much land that they made us work in the fields; and we are what we are, and look what men have become. Among the three sons of my father… (the number of motorcycles, brothers, and brothers’ sons gets confused here, but suffice it to say that her brothers and their children own many vehicles). And now we women have to survive with such great difficulties… Yes, there is a song, something like, to a woman neither the wealth of the maiti is of any consequence

106

nor does she have good times in her husband's house. She has to work very hard in both.

Amongst women involved in agriculture, their hard work and lack of respite reached almost mythic proportions in their speech, although it was mostly couched in modest terms and knowing glances as opposed to boasting. Although she happened not do it in this passage, Shrimati peppered her commentaries on Nepali society, the position of women, and her own life experiences with explicit references to fate and the resignation

with which she faced her lot in life. In her observations that women do not benefit from

economic improvements, Shrimati was referring to her own generation of women. When

talking about her daughters or her future daughter-in-law, she articulated the recognition

that education and economic status would have greater importance for the next generation

of women.

When asked about how her life would change with the introduction of a daughter-

in-law, Shrimati began to joke sarcastically about how a modern buhaari could not possibly bring her any happiness. She explained that she was not educated and barely knew the alphabet, but that her son would need an educated buhaari. Educated women do not do field work. According to older women, today’s young educated women are stereotypically lazy when it comes to household duties and chores. Shrimati said about her son,

(Suppose he) marries an educated buhaari. The buhaari doesn't get a job, and she also doesn't do work in the fields, thinking, “Saasu will do everything, this is enough for me…” Thinking this, she stays in the room with the door closed in the afternoon resting, also not bringing in any income; and about that which my son earns and brings she says, "Don't give to your mother. Don't give to your father. I need it." That being the case, my son will change, then. I also have to calculate thinking about that.

107

Continuing amongst interruptions of poorly suppressed laughter, Shrimati said, “…and

she’ll say to me, ‘Hey, Saasu – make and bring me some tea.’ No, she won't even say

‘Saasu.’ She’ll say, ‘Hey, servant, make some tea and bring it here.’ Isn't it so?!” The

vast gap in the education – and in a related way, the sophistication – of the upcoming

generation of daughters-in-law threatened the old family hierarchies. Thus this kind of

reversal of the domination by mothers-in-law of their daughters-in-law was a typical

subject of the joking amongst women as they looked ahead and envisioned what their

sons’ wives would be like. They were doubtful that a buhaari would bring them rest and

tranquility, yet they thought that it was good for a son and buhaari to live together with

them in a joint family.

Like a potter’s wheel: Women and the family cycle

I have explored the multifarious ways that women’s positions differ across the life

cycle, in the presence and absence of agricultural labor, between joint and nuclear

families, and in respect to historical and social changes by portraying the perspectives of

young wives and mothers-in-law. Now I deploy a different approach to the same topics –

presenting a portion of a single afternoon-long conversation among two generations of

Thakuri women from neighboring joint families.

This household was slightly atypical in a way that was difficult to pinpoint.

Members of this family always seemed to be in flux, in their physical presence in the

home, whether or not a certain person was mentioned as a member, and how their

relation was defined.3 The core members of the family (who I name according to their

3 In order to ensure that the family remains unidentifiable I will not provide details to explain, but a history of two other wives and their children complicated the kinship relations within this household.

108 kinship status) were the aged but able matriarch (Saasu) and patriarch; their eldest son, his wife (Jethi Buhaari), and their three children; their youngest son, his wife (Kaanchi

Buhaari), and their daughter. Only the women and Kaanchi’s young daughter were present that day, along with a neighboring mother-in-law (Neighbor Saasu) also of the

Thakuri caste and a friend of Saasu. Kaanchi was officially the one being interviewed; otherwise she probably would not have taken the lead in responding (evident in the first portion of the transcript) since she was the lowest in status. She gradually grows quiet, until we return to the semi-structured format and the questions of the interview that address her. I give the women’s words, tangents, and explanations precedence, rather than prioritizing my analysis. Instead of selecting and abstracting quotes relevant to my argument, here I give priority to the conversation as it unfolded.

This conversation offers a glimpse into the changes and continuities in gender roles between generations as they are subtly negotiated by junior and senior female family members in one another’s presence. The framing question for the discussion that follows was, “In your opinion, how do women have to be in this society? According to your culture, how do women have to behave?” This question came an hour or two into the interview session, and by this point an air of playfulness had developed that was driving some of the sarcasm and teasing that was typical among the women of this joint family. My research assistant, Meena, capitalized on this lightheartedness and used it as an opportunity to press the women on a few of their responses in the same way that the women gently challenged her and me. The discussion was accompanied by raucous laughter, absent for the most part in individual interviews with women and especially with young women. An element of performance was operating in the discussion that

109

afternoon – a performance by the local women for the outsiders, but also for one another’s enjoyment and to negotiate their respective positions – and it resulted in a comical portrayal of the difference between the paths of men and women.

Jethi Buhaari (JB): How do they have to be?! For some it will be this way, for some it will be that way

Meena (M): Despite that, in society how does one have to be to be a good person?

Kaanchi Buhaari (KB): good…

Neighbor Saasu: Society says to be literate, to study, to be experienced, get a job, earn money, do housework, mind Saasu and Sasuraa, and society will like everyone. If being lazy sits inside the room all the time, [people] will say [that person is] lazy. (laughs)

KB: Around here we have to work, ay?

Neighbor Saasu: And what to do then! We have to work. It's like that!

(at the same time) KB: For doing that they will say "good," and for someone who doesn't do, "bad" Neighbor Saasu: Feeding, giving drink, respecting guests when they come, if everything is done, for that "good" is said. If at the time of guests one sits being grumpy, if there is housework goes into room to sit, what's the use of that laziness.

M: In your opinion, then, if women do all of the housework, society likes them.

KB: Yes.

M: Even if they have studied (been formally educated)?

JB: No matter what.

M: And with sons, even if they don't do any work, it will be enjoyable/fine.

KB: Boys mean boys.

M: Why then?

KB: Don't boys (male) and girls (female) make a difference then?

110

Saasu: The work boys do will be separate. The work women do will be separate.

KB: Women will have housework, boys will have other work.

JB: Boys don't get up in the morning and potne (cover floor with fresh mixture of mud and cow dung – a traditional job of daughters-in-law), it doesn't suit.

M: Why?

Jan (J): (laughing) Why not?

M: Why doesn't it suit them?

J: It doesn't suit them?

Saasu: It also doesn't suit them to wash clothes.

JB: It doesn't suit them, and what to do?!

M: Why doesn't it suit them?

JB: Everything suits us. Being daughters (women). [But] we can't go outside and earn income.

M: Why? You all also go to melaa, do field work…

JB: That melaa… then again, we don't know how to hoe. We only plant, and sons hoe steadily. And boys do the work of boys, and girls do the work of girls.

M: Then again, even if boys don't do work one day, it doesn't make any difference (to anyone). Being a daughter or buhaari, there won't even be one minute without work.

JB: For us it is always the same. Getting up tomorrow morning, that (work); getting up day after tomorrow, exactly that. (laughs)

Saasu: Women's housework means…

JB: Where are the sons that live in the house? Where…

The women subscribe to separate spheres of work and responsibilities – a clear division

of labor – between men and women, but they playfully mock it at the same time. The act

described by Jethi of wiping the mud floors in the morning seemed so utterly feminine

111 that the very image of a man doing such work caused her and the others to laugh. She employs the language of “it doesn’t suit them” as an explanation, but she notes the limits of that explanation by following it with the joke that everything suits women. She ends by pointing out that there were no men in sight at the house. In some cases their absence was because they were working, but even when they are not working men will not stay at home. The conversation continued,

Saasu: After having sons and daughters that do everything, don't we get to sit worry-free? If there had been no sons and daughters, (I) would have had to do even that which (I) can't do, dragging myself would have to do. I have sons and daughters-in-law, grandchildren, and they will do it.

Neighbor Saasu: For that a son and daughter-in-law is necessary.

Saasu: Yes. For that a son and grandchildren are necessary. A buhaari is needed.

M: Meaning, women have to do however much work they get, ay?

KB: Have to.

M: Whether educated or not.

KB: Despite however much they're educated, they have to clean the baby's shit, after having a baby. Who will clean it then?!

M: Husband also shouldn't clean?

JB: Shouldn't do.

M: Why?

JB: It doesn't suit them. (with sarcasm – all laugh)

JB: Do boys wear this (sari or skirt)? We wear like this, and it suits us.

M: These days they wear, they wear it. (joking)

JB: And they don't wear this, boys (gesturing toward her jewelry).

J: Earrings, necklace - they wear.

112

JB: Boys don't wear saris. How could it suit them! Boy and girl were sent differently from up above. And don't we have to distinguish?

J: How did it become that way?

JB: Who knows!

KB: How, how!

JB: How, how. I don't know. From ancient times the traditions have been like this. Who knows, no?

M: But compared to before it has become a little different, probably.

JB: Now a little by educating, listening, it has become (different).

Saasu: Before people…

JB: Now there are educated people that say boys and girl should work equally, no? Boys and girls work equally, eat equally, wherever they go they walk around equally, in some places that is also true. In village households like ours it's not like that. In such a village there will be village households. In the city, the bazaar, outside, then - boys and girls walk together. Both wear pants. Boys and girls, no? Here, if our husbands wear pants, do we wear pants?! Ours is like that.

M: But long ago people used to stay wearing a sari. Now people also wear this much (referring to her lungi skirt and cholo blouse), isn't it so?

JB: Yes

M: Change occurred.

JB: While working in the house we wear like this, if we have to go here and there (means outside of village) we wear sari. We carry our bag, and making ourselves look tip top walk about. Here while doing work with livestock, going to cut fodder, why would we wear sari?! We have to work.

M: Yes. Long ago, after getting married you had to cover your head with your sari, no?

JB: Yes.

M: And that has changed now, no?

JB: Yes. We've already become old, we know everyone, so there is no embarrassment. Long before one had to cover, I've heard, and… (unclear) not

113

show one's head to husband's older brothers. It was like that. We don't have any husband's elder brothers (jeThaju) in our home, then why should we cover the breast that has already been sucked by the baby…

A large portion of the rationale for having children was the need for a son and daughter-

in-law to take over productive tasks once one becomes too aged to do it for oneself. Thus the workings of the domestic cycle and the subsequent need for sons were explicitly talked about in agricultural families like this one. Although Jethi discussed changes in gender roles in cities, she distinguished rural areas like Vishnumati as being a different case, particularly in terms of dress. She identified equality between the sexes as an urban phenomenon. Similarly, she thinks the education of a woman would not mitigate family duties and chores in a household such as hers.

Much later in the interview, she continued with the theme of a sari both representing and hindering women…

For them (men), children don't follow after them. No one gives trouble (to) boys; after saying let's go, they (men) slip on their pants and go. (Laughs.) After putting on sari they (children) grab on. Immediately. Saying, "I will come too!" And where to carry them?! Should I carry on this side, on that side, or walk by myself?! (Mimicking shifting a small child from one hip to the other in exaggerated fashion). What can I do?! (All laugh.) That being the case, I stay at home. Even though I want to go, I stay quietly.

Her Saasu supported her story, saying, “After mothers say they are going somewhere, the children will also say "I will also go" of course. And in a line (they follow after her)…”

Jethi complained that children cannot walk long distances and that she cannot carry them.

Thus she prefers not attempting to go out. Men, on the other hand, “Eat, wash up, and go.” Mobility was thus another major distinction between men and women.

114

The focus then shifted to my situation, and the women teased me by saying that I

should have two children and then do family planning. They also teased Meena about not having to work hard because she was still unmarried and living at home…

JB: …Now because you are still at maiti you are worry-free (to Meena), you are able to go and walk around, no? After returning home, mother and father say, "Daughter, come here to eat," with ease, no? If we walked around all day and returned, (Saasu would say), "Here's your rice" (using "ta" form: a highly disrespectful or “low” verb form).

(laughs)

M: Your Saasu is laughing!

(all laugh)

JB: It is also not her (Saasu's) own home. She left her maiti and came, no? Saasu also like that, buhaari also like that.

KB: [That’s how] it works (chalcha).

Saasu: [That’s how] it works.

Neighbor Saasu: It's like a potter's wheel. It goes around and around.

JB: She also left her maiti and came here. She was not born here. We were also not born here. Leaving maiti one comes. You also, as long as you stay at your maiti, you are so worry-free, no...

In the discussion of gender in this chapter, we now arrive back at the opening topic of women’s positions as daughters in their natal homes. The ease of living in one’s maiti is contrasted with life in one’s husband’s home, but Jethi makes the astute point that the home/household in which she lives is also not her Saasu’s home. Thus Jethi momentarily

unites the distinct and conflicting roles of buhaari and saasu under the broader, shared

category of being a woman in a patrilocal society – of being a woman who does not live in her own home. The domestic cycle, the movement of women at the time of marriage,

115 and the transition of women through familial and domestic roles are all “like a potter’s wheel.”

In this research, the patrilineal, patrilocal features of Hindu Nepali society emerged as some of the strongest and most resistant to change in terms of the factors that influence women’s positions and reproduction. They buttressed existing gender norms and perpetuated son preference, despite dramatic changes in education and urbanization.

As Das Gupta pointed out, referencing Japan and South Korea as examples, “There is little evidence, then, of the quick convergence to the western conjugal family system, which is often expected to accompany the process of industrialization and urbanization”

(48). In my research I could not identify any long-lasting trend towards a nuclear family system, despite a number of de facto nuclear families being present in the study. As I described in this chapter, women in nuclear families almost unanimously voiced a desire for their family to regain a joint status with the marriage of a son. I discuss the implications of this in Chapter 5, describing how the joint family was more significant in influencing fertility behavior as an ideal than it was in practice.

CHAPTER 4

DESCRIBING AND INSCRIBING REPRODUCTIVE BODIES

Cultural norms and structural forces are inscribed upon people’s bodies in

numerous ways. These forces are inscribed on bodies in a symbolic or metaphorical

fashion when the body is considered as a text. Social forces are also inscribed on bodies

in a more literal sense through making visible physical marks on the body – various

forms of body modification including tattooing, scarification, and piercing have been

analyzed in this way. In both of these ways, social norms are inscribed on Nepali women’s bodies through the reproductive processes of pregnancy, miscarriage, birth, and

postpartum recovery. In a symbolic sense, Nepali women’s bodies may be read as a text

of beliefs and values of the greater society, not merely the social expectations for procreation. Furthermore, in a more literal sense, the lines of scars from caesarean sections, tubal ligations, or Norplant use may also be read. The body can serve as a living record of a history of events and conditions that an individual encounters.

In this chapter, I attempt a hermeneutics of the reproductive body in a particular social, historical, economic, and gendered location while acknowledging the forces of

globalization that impinge upon that location. I begin, now, by treating the body as a

text, as a passive entity that is acted upon by and responds to structural forces. But I will

move beyond that treatment of body as text in the next chapter and argue that the

reproductive body is also a source of action, and furthermore, a site of the dialectic

116

117

between structure and agency. The body has become a central problem in anthropology,

and recent scholarship has departed from “the tendency to treat the body as a passive

lump of clay or tabula rasa upon which society imposes its codes, toward understanding

it as a source of agency and intentionality, taking up and inhabiting the world through

processes of intersubjective engagement” (Csordas 1999:178-179).

I follow an interpretive approach only insofar as describing what it is like to be a

social actor of a particular kind, and then I speculate as an outsider on the role of culture

versus economics in explaining models for pregnancy, birth, and the postpartum period.

I offer women’s narratives of reproductive processes and a few of my interpretations of

their situations, but I follow Bourdieu in his suspicion of both local actors’ accounts and

of social scientists’ formal models – realizing that neither can be sufficient. I rely upon

my own situated knowledge as an outsider and a social scientist in my interpretations of

women’s experiences of procreation.

A descriptive overview of maternal health

The experience of procreative processes for women in Nepal varies along the

lines of caste, class, and place of residence. There are dramatic differences in the

practices and experiences of procreation for women in rural versus urban areas and in wealthy versus poor families, and the amount of professional care that women receive during pregnancy and birth has changed noticeably within the last decade. Going to the nearby teaching hospital for antenatal check-ups was unheard of for women in their 30s and 40s in my case studies, but they freely acknowledged that contemporary young mothers would often go. National level statistics from the 2006 Nepal Demographic and 118

Health Survey (DHS) on antenatal care support this finding – Table 4.1 shows a marked increase in the percentage of women in younger age groups who utilize antenatal care

Table 4.1. Antenatal Care Percent distribution of women age 15-49 who had a live birth in the five years preceding the survey by the percentage receiving antenatal care from a skilled birth attendant (SBA) for the most recent birth (2006 Nepal DHS)

Doctor, nurse, or Number of midwife No one women Mother's age at birth <20 50.8 18.6 720 20-34 44.9 24.6 2,971 35-49 20.4 53.4 375 Residence Urban 84.6 12.1 536 Rural 37.3 28.3 3,530 Mother's education No education 28.5 36.3 2,357 Primary 51.9 18.6 743 Some secondary 68.1 10.2 683 SLC and above 89.7 0.9 282 Wealth quintile Lowest 17.7 49.5 956 Second 30.5 28.4 859 Middle 38.4 22.9 811 Fourth 60.7 14.6 752 Highest 84.1 7.6 687

from a trained health professional. This increase is a result of a historical change – an increase in availability of services and the development of a trend in going for antenatal check-ups – and the fact that a higher percentage of women seek antenatal care for their first birth but not for subsequent births. This difference in behavior for first versus subsequent birth holds true for seeking a hospital (or other health facility) delivery for the first child but not for higher parity children (see Table 4.2).

The majority of women in Vishnumati delivered most of their babies at home.

Some women utilized the teaching hospital, which was only approximately 20 minutes 119

away by public bus and able to be reached much more quickly by taxi or ambulance. To my surprise, a few women reported even going to the maternity hospital located in

Kathmandu that was at least an hour away. Going to a private hospital was out of reach economically for the middle and lower class families in my research, but wealthy families in or near urban centers have that option.

Table 4.2. Place of Delivery Percent distribution of births in the five years preceding the survey according to place of delivery (2006 Nepal DHS)

Health Number of facility Home Other births Birth order 1 31.7 66.8 1.5 1,676 2-3 14.8 84.2 1 2,342 4-5 6.9 91.9 1.3 946 6+ 6.3 91.6 2.1 580 Residence Urban 47.8 51.5 0.7 677 Rural 13.5 85.1 1.4 4,868 Mother's education No education 7.9 90.7 1.4 3,343 Primary 18.9 80.2 1 1,009 Some secondary 34.6 64.1 1.3 848 SLC and above 67.4 31.6 1 345 Wealth quintile Lowest 4.3 93.3 2.4 1,412 Second 9.3 90 0.8 1,180 Middle 11.9 87.1 1 1,132 Fourth 21.7 77 1.3 983 Highest 55 44.3 0.8 838

At the national level, the large majority of Nepali women give birth at home. In

the five years preceding the 2006 DHS, 81 percent of births took place at home. A total

of 18 percent of births were delivered in a health facility, and less than 1 percent of births

occurred in a private health facility. The majority of women who had given birth in the

five years before the survey reported that they believed it was not necessary to give birth

in a health facility (73 percent), 17 percent said that it was not customary, 10 percent 120

cited that it cost too much, and 9 percent said that a health facility was too far or that there was no transportation available. In addition, 3 percent of women mentioned that

they gave birth before they could get to a facility even though they had planned to go to a

health facility for delivery (Ministry of Health 2006).

Nepal’s maternal mortality ratio of 281 (per 100,000 live births) for the period

1999-2005 (Ministry of Health 2007) makes maternal health a concern of the Ministry of

Health and non-governmental agencies alike. Hygienic conditions and the assistance of a

trained medical professional during delivery can reduce the risk of complications and

infections that may cause serious illness or death. Urban residence, a higher level of

education, and a higher level of wealth all corresponded with greater utilization of

delivery care by health professionals in Nepal (see Table 4.3). Several factors have

Table 4.3. Delivery Assistance Percent distribution of births in the five years preceding the survey by person providing assistance at time of delivery (2006 Nepal DHS)

Doctor, nurse, or Relative or Number of midwife other* No one births Mother's age at birth <20 22.1 46.1 2.5 1,156 20-34 18.5 50.3 6.5 3,957 35-49 11.3 53.2 17.8 433 Residence Urban 50.6 32.7 2.9 677 Rural 14.3 52.0 7.0 4,768 Mother's education No education 8.2 55.8 8.8 3,343 Primary 20.6 49.7 5.6 1,009 Some secondary 36.6 38.7 1.1 848 SLC and above 71.1 17.1 1 345 Wealth quintile Lowest 4.8 68.2 13.7 1,412 Second 10.1 52 4.1 1,180 Middle 12.4 45.3 6.2 1,132 Fourth 23 44.5 3.2 983 Highest 57.8 26.9 2.2 838 *“Other” refers to any non-trained person, such as a neighbor 121

contributed to the decline of maternal mortality in Nepal: an increase in the percentage of women making four antenatal care visits, the percentage of births delivered at health facilities, the percentage of births assisted by a skilled birth attendant, and the percentage of women receiving postnatal care (Ministry of Health 2007). The identification of

“risks” and the Foucauldian surveillance of beliefs and practices by the governmental, local, and international agencies are clearly both at work here: in the nature of the creation and administration of the DHS and in the goals of the government to reduce maternal mortality. Additionally, through medicalizing pregnancy and birth, a further form of biomedical surveillance is enacted through antenatal check-ups and delivery assistance. It is assumed that such surveillance of citizens and bodies, and the resulting social and biomedical interventions, will serve to alleviate suffering and mortality related to procreation; but that assumption does not go completely unexamined in the following discussion.

Hidden pregnancies

Becoming pregnant is perhaps the greatest duty of a new wife. Married women explained that if a woman does not become pregnant within two years after marriage, gossip will start in the community and the family will begin asking questions about what is wrong. A woman is highly valued within the ghar for her role as producer of children for the family. While pregnancy is eagerly anticipated within the family, at the same time it is not a topic that is discussed openly. 122

Announcing one’s pregnancy to family members, either to one’s husband or

parents-in-law, was an idea that most women scoffed at or recoiled from in

embarrassment. Women pointed out that the family members will almost certainly notice

within the first few months that a woman’s menstruation has stopped – the husband

through intimate contact with his wife, and the parents-in-law through the absence of a

woman’s monthly untouchable status (the most obvious and most followed observation

of the na chunne status is the few days when a women will not cook rice for the family).

Another common way of finding out, both for themselves and for others, was when

women had difficulty eating dhaal bhaat (the staple, ubiquitous meal of rice, lentil soup,

and vegetable curries and sometimes meat) or cooking because of the smell of vegetables

or dhaal. Some women described not knowing what was wrong with them when they

became pregnant for the first time, but they remarked that these days young women will know, even in the village.

Some bodily experiences blur the symbolic and physical distinction I made at the

outset of this chapter, such as the effects of poor nutrition on the mother and developing

embryo during pregnancy. In a subjective fashion women experience the effects of poor

nutrition through night blindness, weakness, and dizziness. Several women described the

problem of nausea and not eating in terms of a cycle. A mother of two in her 30s

described how after the nausea started and she did not feel like eating,

Just like when [one is] sick, arms and legs feel exhausted, and one wants to keep sleeping, that’s what happens. The person will become thin, won’t want to eat, wants to spit, nausea comes regularly, and dizziness from time to time; I discovered it is like that. If one becomes weak while pregnant, then the feeling of not wanting to eat is even stronger. Sometimes after sitting and when about to get up to do some work, [I] feel dizzy and can’t see – [I] see black and blue. And because of dizziness I have to sit. After a few moments, after half an hour it 123

became ok. While being weak it will be like that. In the beginning (of the pregnancy) it was like that.

Such experiences do not leave lasting visible marks on the body, but one experiences the

subjective effects of poor nutrition in the ways described. The woman quoted above also

described the observation of her first baby being “born with his stomach stuck to his

back” and how the hospital attendants scolded her for not eating enough during pregnancy. In this way there was visible evidence of her not eating, and an opportunity for reflection on that evidence when the hospital staff connected the newborn’s concave stomach to her lack of intake of nutritious food during pregnancy.

Most women reported eating the same things during pregnancy as they usually did

– whatever was available to the household depending on what was in the garden or the amount of money available to purchase things like chicken, goat, or vegetables trucked in from surrounding areas. Some older women joked about having to eat dhido, a sticky mass of ground wheat (or corn or millet) with little appeal in terms of taste or texture, because rice was not always available. Women did not report eating smaller portions of choice foods like meat or vegetables than the others in the family, but they also did not

receive special nutritional treatment during pregnancy. The exception is during the ninth

month of pregnancy when traditionally a woman’s natal family brings her yogurt, beaten

rice, and sometimes fruit and sweets. Several women cited a long distance as a reason why their natal families did not fulfill this obligation.

As for work, many women reported that their household and agricultural duties did not change after they learned that they were pregnant. One elderly woman made light of the whole affair of birth, saying, “If tomorrow I was to give birth, up until today I would be carrying a load of firewood and fodder from that forest (pointing several miles 124

up the mountainside), there where it is foggy, I would reach there.” I hesitate to call this

woman “elderly” since that word often carries negative connotations of weakness, and one day she bet that if she wrestled my female research assistant she would win (my money also would have been on the elderly woman). She looked fit and strong. She claimed that it was easy to give birth, and that other than at the time of her first son’s birth, she was only in labor for an hour each time. Trying to boast and make us laugh, she stated dryly, “I gave birth like defecating.” Regarding the relationship between working and carrying loads and giving birth, she claimed that, “However much work is done, it becomes that much quicker.”

It quickly becomes apparent from similar comments made by the older generation of women that birth is not something about which to make a fuss. Surprisingly, though, this particular matriarch, along with the men of the household, acted with much care and

consideration toward her pregnant daughter-in-law. Her daughter-in-law, Kalpana,

reported that during her pregnancy, her mother-in-law and her husband would go with her

when she climbed and wandered the mountainside looking for fodder to cut for the livestock. She would collect the fodder, but her husband or mother-in-law would carry the load. During that time she was also excused from washing the family’s clothing – she only washed her own. Kalpana’s case and her family were exceptional among my case studies, but they were an important example of the potential for variation. This was also the only household in which special foods like fruit and yogurt were bought particularly for the pregnant daughter-in-law throughout her pregnancy.

Many women engaged in the same level of physical labor during pregnancy as they usually did. Most explained their continuation of physically demanding work in 125

terms of necessity. One mother-in-law stated in a matter of fact way, “Who would do the

work if I didn’t do it?” She had no children old enough to help her, and her mother-in-

law was living with another relative during the time she gave birth. She asked, “How

could I get food to eat without doing work? I also had livestock and fields. I had to cut fodder and cook… bring water… When pregnant, if possible the person will rest, if not – what to do?” Only the very young mothers or those who had higher levels of education had heard messages such as one should not carry a heavy load during pregnancy. One low-caste, impoverished woman was employed to wash clothes during one of her pregnancies, and she began to miscarry while she was carrying a large bucket of wet clothes up several flights of stairs. Only afterwards did she learn from others that the

miscarriage was probably related to her work.

On the whole, it seemed that the time of pregnancy was not conceptualized in

local systems of knowledge as a period of vulnerability for the well-being of the mother

or the future child. Hospitals and health posts recognized this and were involved in

educational campaigns on proper nutrition and wrote prescriptions for vitamins at those

locations. As numerous failed public health measures have demonstrated, education is

insufficient if there are economic or other structural barriers to women receiving proper

nutrition and reduced physical work loads. Thus I hesitate to blame a cultural construct

or understanding of pregnancy as a stigmatized state that one should modestly conceal for

what may be the reality of a shortage of nutritious food in the family as a whole or a lack

of a replacement for a pregnant woman in doing crucial productive household or

agricultural work.

126

Home births and hospital emergencies

As described in the introduction of this chapter, the majority of Nepali women

continue to give birth at home. When considering the socio-historical conditions of ten to twenty years ago, the reasons for this are clear. Vishnumati, for example, was still a rural village at that point and lacked paved roads and electricity. Women had always given birth at home prior to that point, for there were no other options (except for the urban elite). A woman who gave birth around 1992 described how when the placenta would not come out she and her newborn had to be carried on a man’s back in a dhoko (a large basket used to haul things like fodder or bricks) filled with straw down the mountainside until they reached a taxi stand. Another woman told how after being in labor for three days, her father-in-law walked down to find a taxi to take her to the hospital. During the time he was gone, she gave birth successfully. She said that at that time, around thirty years ago, there was no tradition of going for antenatal check-ups. She said, “And at that time, just barely, the practice of taking women to the hospital for birth had started. As much as they could, women would give birth at home – but rich people could take the woman to the hospital if the birth became difficult.” In her time, cars and taxis were not available at all – one had to walk to a nearby boarding school in order to access a phone.

Then one could call the hospital to send a car.

Although these stories struck these women’s teenage daughters as completely foreign, such scenarios would not be uncommon in rural, remote areas of contemporary

Nepal. Infrastructure is poor in the mountainous regions, and progress in this realm was damaged by the Maoist insurgency between 1996 and 2006. Part of the reason why at the national level the majority of Nepali women give birth at home, then, is because almost 127

80 percent of Nepal’s population lives in rural areas where the availability of delivery

services is unlikely. In the semi-urban village of Vishnumati, however, some other

factors are causing women not to give birth in hospitals or to even call a Skilled Birth

Attendant from the nearby Health Post at the time of delivery.

Most of the women in the case studies who gave birth at the hospital did not speak poorly of the facility or treatment received there. One woman’s experience, however, was harrowing. Shanta gave birth to her eldest child, a son, at the teaching hospital. The

nurses showed her son to her, but then later another nurse picked up her baby and handed him to another woman who also had given birth. Shanta had heard of such a thing happening before, and so she was worried and was keeping watch over her son despite feeling weak. She told, “A woman near my bed had given birth to a daughter. That woman was claiming my son, saying this child is mine. Of course, she was unaware of the switch. I got up and said ‘That is my baby.’ I scolded that nurse. My son would have

been exchanged if I had not taken care! That woman also could have taken my son

because she already had four daughters.” This near swapping of infants occurred after

she had already experienced another mishap. She had needed stitches after giving birth,

and the nurses who were working on her were students at the teaching hospital. She

needed six stitches, but some of the stitches were so twisted that the senior nurse scolded

the student nurses when she saw their work. She ordered them to repeat the process.

After the senior nurse left, the student “stitched it her own way” according to Shanta, and

it started to bleed. Such experiences, even if they are rare, do not build confidence in

hospital deliveries amongst women or their families. 128

Birth was predominantly viewed in Vishnumati among the middle and poor

classes as a natural process. It had not become medicalized; it did not require the

expertise or assistance of a trained medical professional. Hospitals, for the most part,

were perceived as being for health emergencies.

Women were socialized to keep quiet about their suffering, and it was men who

made decisions such as determining at what point the situation was dangerous or life- threatening enough that it warranted being taken to the hospital. The way in which a young mother in her 20s told the story of the complications that occurred after she went into labor and the actions that were taken provide insight into the different roles family

members act out in such scenarios:

Q: How long did you have labor pains?

A: I had labor pains from Friday evening, and I gave birth on Sunday at 11:00 a.m.

Q: When did you go to the hospital?

A: I went on Sunday. I did not agree to go, but later elder brother and others scolded me saying go to the hospital. They said that they would take the baby out by doing operation or something good. And I said I wouldn’t go. Elder brother (in-law) kept scolding, saying that I had to go. Elder brother (in-law), mother (in- law), husband, the three of them, and one neighbor, four of them, took me to the teaching hospital. I was checked at the teaching hospital, and then I was taken to the maternity hospital in Thapathali in order to give birth.

Q: Why?

A: Brother (in-law) said that it is better at Thapathali than at Teaching. So I was taken down there.

The young woman refused to go to the hospital initially, even after being in labor for two

days. She made a point of saying that twice she refused, and then finally she gave into

the reprimands of her husband’s family. At another point in the interview just after this 129 story, she claimed that the labor pain was not that bad – she had “slight stomach pain.”

She deliberately minimalized her role and her agency in this situation in order to follow her understanding of the social script of being a good woman and daughter-in-law. She did not want to appear to be demanding, even when she was telling the story with only me and my research assistant present. After she gave birth at the hospital she lost much blood and fluids, and she remained there for four days on an intravenous solution.

In addition, women are often in no condition to demand to be taken to the hospital

– they may be barely conscious during an obstetric emergency. Shanta delivered her second child at home. There was “a lack of money” at that time, and since she started to have labor pains around midnight she thought, “Why give trouble to others in the night. I called to my mother-in-law just before she was born.” She figured that there was no one to take her to the hospital in the middle of the night anyway, so why should she wake the other family members. The labor pain was not as bad as it was with her first birth, her son, and she gave birth “easily” to her daughter around five in the morning. Shanta said,

She was born easily, but the placenta did not fall easily. It did not come out for 2- 3 hours, so it became difficult to cut it. My daughter became so ‘serious’ and so did I. Different people were saying different things. Whom should I believe? I was ‘serious’ because of bleeding, I was in a dilemma… what to do? I fell unconscious for about twenty minutes. All the family members were weeping. They thought I was dying.

Shanta did not indicate how much time passed in this way, but she said that the family members called the village midwife. The midwife was able to pull out the placenta, but there was excessive bleeding. The midwife recommended that she be taken to the hospital. Someone fetched a taxi, and they took her to the teaching hospital. She regained consciousness on the way to the hospital. 130

Shanta remarked that many women had died when the placenta did not come out.

Some old people, she reported, had said that cow dung should be thrust in her mouth –

others said hair – so that she would vomit. Some recommended using a small hoe, and

Shanta winced in telling this, citing the possibility of tetanus. She had received some

training as a volunteer village health worker, and she knew what should have been done

in her situation according to a biomedical model – but she said she was unable to speak at

that time.

These two stories and other similar ones highlight the importance of male

involvement in obstetric emergencies in this social context. Women are likely to be

limited, for both social and physical reasons, in their ability to decide that their situation

necessitates hospitalization during obstetric emergencies. Part of the breakdown in

women giving birth in health centers or with the assistance of a trained professional, and in women being hospitalized in a timely fashion during an obstetric emergency, may result from a combination of factors. Birth is considered a natural event that does not warrant much special attention. Older women told stories of giving birth alone during the night or on the way home from the fields. In the past, a female relative would help a woman during delivery. Birth was both the domain and responsibility of women. With the advent of hospital deliveries and availability of trained health professionals to assist with birth, obtaining such care began to involve the decision-making power and the initiative to act that was in the hands of the men in the family.1 But men have not been knowledgeable about birth in the past, and may not even be alerted that it is happening.

1 Carolyn Sargent made a similar point about the medicalization of birth among the Bariba of Benin: she took the argument a step further by concluding that women’s reproductive choices were being limited by the encouragement of hospital-based births by handing control over obstetric care to men. 131

In a much different setting in Nepal, Mullany’s ground-breaking research with

couples delivering in the major public hospital in urban Kathmandu revealed that a key

obstacle to Nepali husbands’ involvement in antenatal care and deliveries was their lack

of knowledge regarding women’s maternal health, along with social stigma and shyness

or embarrassment. Mullany concluded, “Appealing to men as ‘responsible partners’

whose help is needed to reach the endpoint of ‘healthy families’ may, for example,

provide an effective approach for targeting men in the Nepal setting” (2006:2808).

Young men and health providers alike in her study stated that young men (in this setting)

were ready to be more involved with maternal health, but they needed education and

either the will to ignore the social stigma attached to helping one’s wife or programs

aimed at changing such stigma.

In places like Vishnumati, a different sort of male involvement is needed. A

disconnect exists between the risk for female bodies and the male involvement in mitigating that risk. Increasing the number of births delivered by a trained health professional may require efforts to make safe births and healthy infants the responsibility of men. The range of the ideals and responsibilities of fatherhood needs to be extended

to include active preparation for birth, whether at home or in a hospital. At the national

level, few people made specific preparations before delivery (of the most recent birth in

the five years before the survey): 46 percent of women made no preparations, 29 percent

of men made no preparations. Women’s efforts were concentrated on saving money and

arranging for food and clothing for the newborn, whereas men focused on saving money

(see Table 4.4). In a similar fashion, educational campaigns about the importance of

good nutrition during pregnancy and advice about what work should be avoided would be 132

Table 4.4. Birth Preparedness: Women and Men Percentage of women and men who made specific preparations before delivery of the most recent birth in the five years preceding the survey (2006 Nepal DHS) Bought Arranged Found Contacted clean Food Saved for blood health delivery and No money transport donor worker kit clothing Other preparation

Women 37.1 1.4 0.3 3.5 9.1 26.2 3.9 45.8 Men 53.9 5.5 1.0 8.7 9.9 N/A 3.8 29.3

helpful. Most women are able to continue to work during pregnancy, but a woman and

her family members should be knowledgeable about what kinds of physical activities are

most risky for the mother and fetus. That way other children or relatives may be

expected to help with particularly risky tasks, at least in scenarios where such help is

available. The success of immunization campaigns and the promotion of contraception at

the national level lead one to believe that interventions in terms of promoting maternal

health could be successful.

Protected postpartum periods

The inscription trope is a familiar one in anthropology, but it is of interest that

women themselves talked in a similar way about how certain “unhealthy” actions during

the recognized postpartum period (sutkeri) could have long-lasting or permanent negative

effects on the body, particularly in the form of pain. Unlike during the periods of pregnancy and birth, the bodily vulnerability of a woman is explicitly recognized during the postpartum period. During this culturally defined period of seclusion, women usually are not allowed to cook or do any household work other than to wash their own clothes

and dishes. Feeding women larger amounts and certain types of foods are also

emphasized during sutkeri. 133

Sutkeri is both a state of being and a period of ritual seclusion for a woman after she gives birth that lasts for at least eleven days. Birth is considered a polluting process, mainly for the mother but also for those involved in the delivery and even the household in general. Traditionally a woman is considered untouchable until she “touches water” on the eleventh day after birth. On the eleventh day a priest a conducts a ritual of purification and a naming ceremony called nwaaran. Sometimes a woman will not

“touch water” until the twenty-second day, in which case she remains in a polluted state

and is not allowed to cook, enter the kitchen, etcetera.

Women expressed differing opinions about exactly when this period of pollution

begins – some women thought that after the sixth month of pregnancy women should no

longer conduct the household puja or go to the temple, and some thought that such

avoidances should begin with the discovery of pregnancy. Pregnancy, however, was a

dramatically less impure state in women’s descriptions than sutkeri. Much more

emphasis was placed on the rules governing sutkeri than on pregnancy because the

process of birth and the bodily fluids involved confer greater impurity. After “touching

water,” a woman is purified and may return to her regular activities; however she should

not do puja until the first feeding ceremony of the child (paasni), which occurs after 6

months for a son and 5 months for a daughter. In practice there are exceptions to these

rules, as dictated by economic necessity or by a family no longer buying into this system

of beliefs. One good example of the contingent nature of these strictures came from a

young low-caste woman living in a nuclear family with her husband – after giving birth,

her husband massaged her body with oil daily. Such a blatant disregard for her state of

impurity by her husband would have shocked most of the women represented in my case 134

studies, but she simply explained that he would bathe after touching her and thus resolve

the issue of pollution.

In a similar fashion, women related the speed with which they returned to work after giving birth to a combination of economic and familial circumstances, and

occasionally to one of the following: being educated, being young, or being modern

(which are all somewhat related in this context). Most women agreed that for the first

eleven days, the only work a woman should do is to wash her and her infant’s clothing

and plates, these items having become jutho (ritually polluted) through contact with the

mother. Then, as one woman explained succinctly, “After eleven days, if she has people

to do the work at home then she will not do a thing; but if there is no one to do the work,

she will do everything herself.” And in some cases, even if there was a mother-in-law

there who could assist with household tasks, one young illiterate woman said, “Why

would she help?! While there is a buhaari, why would mother-in-law do it?!” She had to

do the same work after giving birth that she had done before after the 11 day period of

seclusion. For women who were involved in agricultural work or owned livestock, this

likely involved physically demanding work such as carrying loads of fodder in a dhoko or

bending over to plant rice or wheat. Possibly more than any other task, fetching water

from the tap is considered women’s work. Kalpana’s mother-in-law, the matriarch of the

exceptional family mentioned earlier, knew the risks of carrying heavy loads and thus

developed a strategy for not carrying a heavy water pot. She carried a smaller vessel to

the tap so that her load was lighter. Most young mothers no longer faced this problem,

for indoor plumbing was becoming more common and was standard in the concrete

houses in the bazaar area. 135

Women whose maiti was nearby went there after purification, spending anywhere from a month to six months in the care of their natal family. Women “don’t wash a single plate” when they visited their maiti – they were pampered by their mother and family. In the end, women referred to “sutkeri” as the total time they spent resting, often a combination of time spent in seclusion and then at one’s maiti. The jutho portion of sutkeri was defined as only the first eleven days or alternatively the first twenty two days of sutkeri, depending on when a woman was purified.

The bodily practices prescribed during sutkeri are a mixture of responses to this state of pollution and also the vulnerability of the mother’s (and infant’s) body to cold, water, and air through openings in the body and the ingestion of substances with certain properties. For example, foods with a heating property are fed to women during sutkeri, and sutkeri women are supposed to avoid eating food with cold properties (chiso laagne) such as fruit (except mango), kaalo daal (black lentils), cauliflower, eggplant, curd, and gourds. Women were also traditionally fed ghee and chaaku (molasses), both with warming properties, and large amounts of rice. Depending on whether the family is vegetarian or not and whether such items can be afforded, families feed sutkeri women chicken, fish, or goat. Women avoid chilies and leafy green vegetables for fear of causing gastric trouble for the infant, but a few types of leafy greens were acceptable.

After giving birth women are fed jwaanoko jhol, a soup made from jwaano seeds (similar to cumin), which contain calcium and magnesium (Moser et al 1988). This concoction is reported to help women produce sufficient breast milk. A few women also mentioned another medicine often made at home for lactating women from ground spices, and 136

sweets made with cashew, molasses, raisins and other dried fruit are a common gift for

new mothers.

Other sutkeri practices also involved keeping the woman and infant warm. Many

women reported building a fire inside the room of the house where the sutkeri woman

was staying, and keeping the windows and doors closed to prevent a draft. Some women

took these precautions even during the hot months prior to the monsoon season; others at

least opened the windows. Most women wrapped their heads and waists with cloth while

sutkeri. Tying a piece of cloth around the head helped prevent headaches and prevented

cold air from entering at the top of the head. Some women said that they felt cold if they

did not wrap the head and keep their chest covered. During sutkeri women wear a

choubandi cholo, a long-sleeved blouse that ties in four places in the front and covers the

chest with a double layer of cloth. Women wrapped their waists for a couple of reasons –

mainly so that air would not enter and the stomach would become small again, but also as

a support that prevented pain after doing work.

Women massage themselves and the infant with oil, usually twice a day if

possible, once in the sun and once in the warmth of a fire. This practice was also explained in terms of preventing cold and air from entering the body so that the body would not swell. There was also a concern of water entering the body, particularly

during bathing, and the generous amounts of oil used on the body served to prevent that

from occurring. Women should not bathe until the tenth day after birth before the

purification ceremony on the eleventh day. Similarly, women are not supposed to comb

their hair or brush their teeth until the tenth day. Brushing with salt and oil or pieces of

coal was acceptable, but using a toothbrush or toothpaste was not allowed. The teeth are 137

said to be weak during this time, and for the same reason women do not eat beaten rice

(which requires much effort to chew). As one can imagine, not bathing for ten days

while massaging the body with oil leads to a sticky, smelly mess – and the smell of a sutkeri woman is the source of much joking. One woman joked that she almost died from being so disgusted by her dirty condition.

Naturally not all women followed the rules for sutkeri, especially those who gave

birth during the hottest months. One woman could not tolerate not brushing her teeth, and another admitted to combing her hair – but she changed her response after her mother-in-law heard her and cried out, “You combed your hair?!” Several women did not know the rationale behind some of the practices, so in households where there was no older generation to enforce such rules they often fell by the wayside. One young low- caste woman who ignored many of the sutkeri rules and returned to her work of cutting cloth and sewing was scolded by others. They told her that her body’s condition would get worse later, saying, “What do you know now? You will know later.” The effects of not taking good care of the body during sutkeri were considered to be long-lasting but not immediately felt – they would be experienced in terms of weakness and pain while working later in life.

Women thus described the postpartum period as a time of physical vulnerability that required taking extra care to protect the long-term well being of the body. They talked about sutkeri as a period of rest and healing, and a time when their regular duties were suspended. No one depicted it in terms of unwelcome seclusion or hardship, other than the dislike of being dirty and smelling badly for the first ten days. Critics of the ritual seclusion of Nepali women after birth are accurate insofar as the dangers of 138 infection that exist when women give birth in highly unsanitary conditions such as animal-sheds (see Thapa et al. 2000), but the version of sutkerki practiced in Vishnumati was portrayed by women as a welcome period of recovery.

By and large, I present a model of procreative processes that describes a hidden and stigmatized pregnancy marked by minimal social recognition in contrast to a protected postpartum period denoted by ritualistic and practical social practices. One interpretation is that pregnancy represents a woman’s sexuality – her pregnant stomach is evidence of her threatening and stigmatized sexuality to her husband’s family. The process of birth transforms her from a hyper-sexualized young woman into a moral mother, a contributor to the patriline, and thus acts as the righting or moral correction of her sexuality into the product of a child.

One problem with this interpretation is people’s seeming lack of concern or focus on ensuring that birth is successful. Many women die during or shortly after childbirth, as do infants. If at least the infant (if not the mother) is so important to the patriline, why do people not go to great lengths to protect its entry into the world? Unlike the mostly invisible effects of difficulties during pregnancy (excluding miscarriage – although it is also hidden) that may not be easily connected to negative outcomes in terms of infant health, complications at the time of birth are easily observable and are life-threatening. I propose that the women in this study were caught in a transition between a tradition in which birth was located securely in the realm of women and the category of natural and mostly uncontrollable events, and a movement toward hospital births in which the concepts of risk maintenance, planning, and control are introduced. Given the current 139 gendered aspects of decision-making at the household level, this brings birth squarely into the province of men.

The trope of inscription advanced in this chapter is, in many ways, insufficient. It inevitably positions women and their bodies in a passive mode – as receptors, or texts – on which social norms are inscribed. The metaphor does not allow for a mutually influential relationship, or agency, on the part of women or their bodies. In the next chapter I seek to de-scribe women’s bodies, leaving the metaphor of the text and showing how the body can be a locus of agency and a means of expressing it – whether or not the expression of that agency appears to be a form of resistance to dominant discourses.

CHAPTER 5

CONFLICTING DISCOURSES AND AGENTIVE BODIES

Contemporary young Nepali mothers are positioned amid conflicting trends of fertility decline and persistent son preference. According to my informants, the phrase

“son preference” overstates the situation; women do not prefer sons over daughters,

rather they simply need to produce a son. But in the context of dramatically lowered

fertility in comparison to the previous generation, the pressure to produce a son has become more acute as young women need to give birth to a male in fewer overall attempts. Women’s bodies are located in the middle of these conflicting currents, and their procreation is a site of social continuity and change.

In Nepal, notable increases within a single generation in the education of women and their potential economic contributions to the household would seemingly lead to a decrease in son preference. Using an in-depth discussion of a few case studies with the larger study as an informative backdrop, I examine why son preference persists in such regions despite changes in gender norms and the breakdown of traditional cultural and religious explanations for the need for sons. I consider the role of the joint family system in this context since aspects of kinship or family systems, such as norms for marriage, inheritance, and living arrangements, have bearing on demographic outcomes (Dyson &

Moore 1983; Das Gupta 1997; Skinner 1997).

140

141

A recent comparative project on South Korea, China, and India – all patrilineal,

patrilocal societies – underscored the important role of a similar kinship system despite

local cultural differences in perpetuating son preference (Das Gupta, Zhenghua, Bohua,

Zhenming, Chung, & Hwa-Ok 2003). Likewise, evidence within a single nation of

significant changes in cultural norms and gender roles that have had limited effects on

son preference would support their conclusions regarding the importance of kinship

systems. Building on that assumption, this research examines cultural changes in Nepal

related to son preference, analyzing which cultural norms are questioned by contemporary couples and which remain intact.

Individual negotiation, agency, and variation within the kinship system, including

the differences found in nuclear households, are included in order to avoid the reification

of the extended family system, and to emphasize that part of its power lies in its existence

as an ideal. In this way I captures the dynamic, living aspects of family systems (see

Collier & Yanagisako 1987) rather than being beholden to static models of kinship of the

past that were sometimes equated with “culture” itself in demography (see Kertzer 1995).

Along similar lines, other structural forces such as gender and reproductive norms are

treated as forces that may limit, but not dictate, the range of human behavior. Examining how people negotiate societal expectations through their relationships with others reveals how through the process of everyday dealings people manipulate, tear down, or reinforce norms and larger institutional frameworks (Bourdieu 1977; Ortner 1984).

Young married women’s conflicted feelings about the need to produce a son are presented here within the context of fertility decline and the discourses surrounding it, and simultaneously within and against the older generation’s explanations for the need

142

for sons. Instead of the commonly told narrative of South Asian husbands and mothers-

in-law pressuring young married women to produce a son, this research reveals a more

complex picture of women in this setting internalizing the need for sons, recognizing the

practical reasons behind that need, and acting accordingly. Women act pragmatically in

matters of procreation within the current socio-cultural climate, but all the while they

proclaim they do not need a son.

The unlikely “Happy Family”

In countries around the world, fertility transitions have been associated with economic growth and social change. The onset of fertility transition has followed “the

conversion of economies into ones where practically all transactions are monetized,

where children go to school, where modern medical facilities are easily available, where

an increasing number of mothers received trained health professional assistance when giving birth, and where infant and child mortality rates are falling to at least moderate

levels” (Caldwell 1998). In such an environment, children become expensive and the

allocation of resources within a family shifts (Caldwell 1982). This ideology itself – the

whole-hearted adoption of it and the translation of it into funded programs by Nepali

policy-makers – may have influenced fertility decline in Nepal as much as the actual

conditions described by it.

Despite the country’s status as one of the world’s poorest and the challenges of its

extreme topography to development, Nepal is undergoing a fertility transition.

According to the 2006 Demographic and Health Survey (DHS), the infant mortality rate and the under-five mortality rate, though dropping, remain high at 48 and 61 (per 1000

143

live births) respectively. Acute respiratory infections and diarrheal diseases are the most prominent killers of infants. The large majority of women (81%) gave birth at home

between the 2001 and 2006 DHS, and not necessarily because a hospital was unavailable.

I found that even families who live within a short ride to a hospital often give birth at

home. Despite high infant mortality rates and problems with both access and quality of

care for women during delivery, the total fertility rate declined from over 6 births per

woman of reproductive age in the 1970s, to 3.1 in 2006 (Ministry of Health et al 2007).

Nepal is a nation of disparities, and this is reflected in fertility differentials. There

is a small educated, wealthy, and worldly elite class on one end of the spectrum, and a

rural and impoverished majority on the other end. The lives of these two extremes are

utterly different, and the study population addressed in this research represents the middle ground. Much of the middle hills and the Tarai (the southern plains) regions are becoming part of a global economy and society, and Kathmandu and other urban areas are almost fully so. The mountainous region that stretches the northern length of the country offers a formidable challenge to the development of infrastructure and the provision of services like education and healthcare, or even roads. Consequently, fertility rates are lower in urban areas, which are concentrated in the middle hills and Tarai, among the well-educated, and among the wealthy.

Contributing to fertility decline is the marked increase in the use of temporary methods of contraception between 1996 and 2006 (see Figure 5.1). Historically and presently sterilization has accounted for the largest share of contraceptive use in Nepal, but originally this may have been attributable to the emphasis placed on this method by family planning programs and subsequently local practitioners or field workers.

144

Although it is popular, permanent contraception is used (primarily by women) with

caution as the experience of child death is not uncommon. The age at marriage has also

increased, but the median age at first birth has changed very little in the past two decades

– remaining at 20 years of age (Ministry of Health et al 2007).

Figure 5.1. Trends in Modern Contraceptive Use among Currently Married Women, Nepal 1996-2006 (2006 Nepal DHS)

1996 15 2001 2006 10 Percentage of married women 5

n s m ill le o p UD b d I a n lizatio o c implants inject steri sterilization le le a a m fem

The pace of economic development has been slow for Nepal, and has been impeded further by political events, civil unrest, and the intensification of the Maoist insurgency between 2001 and 2006. The violence and fear caused by the insurgency and the army’s response also increased the divide between life in urban and remote rural areas. Most of the fighting took place in the rural countryside, and a substantial number of men fled to urban areas and sometimes abroad in order to escape the violence and find employment. The drop in fertility documented by the 2006 DHS may be driven, at least

145 in part, by the resulting separation of couples or the movement of couples from rural to urban areas.

With opportunities for education, employment, and new forms of leisure, the costs and benefits of childrearing change and individuals are exposed to new ideas that may influence fertility behavior in Nepal (Axinn 1992; Axinn and Yabiku 2001), as well as across many societies (Thornton and Fricke 1987). In semi-urban areas like the one described here, the transition away from agricultural labor as the main means of production towards small businesses or various wage-paying positions is well underway.

People recognize the value of education, plus the costs associated with it; although they often express discouragement over education not resulting in employment.

In summary, the total fertility rate for women of reproductive age in Nepal fell from around six children per woman to three within two generations. The major cause of this fertility decline appears to be the promotion, availability, and acceptability of contraception and related services. The question of why lower fertility and the utilization of contraception became acceptable remains. Along with Pigg, I argue that Nepalis experienced modernity through a development ideology (Pigg 1996) – in the case of procreation, through a development ideology that was explicitly presented to them by the state.

Historically, the acceptability of limiting family size was driven by a dramatic expansion of the role of the state into everyday affairs, in particular into Nepali’s reproductive lives. This began in the late 1950s with the establishment of the first family planning service organization, the Family Planning Association of Nepal. Around the same time, the government of Nepal began to associate the role of family planning with

146

national development and family welfare. The government concluded that reducing the

national fertility rate would help maintain a balance between population growth and

economic growth, and so it adopted a family planning policy in 1965. Subsequent

development plans dealt with population processes from both a policy and programmatic

point of view. From 1985-1990, population policies and programs not only emphasized

family planning issues in the short term, but also focused on long-term concerns such as

encouraging the small family norm through education and employment programs that

aimed to raise women’s status and decrease infant mortality. The government altered its

position from encouraging family planning through sterilization to promoting temporary methods. The Eighth Plan (1992-1997) continued the strategy to “create atmosphere for the small family size norm of two children through economic and development programmes” and improve the scope and quality of family planning services offered throughout the country, but it also established a new emphasis on the concept of birth spacing and the use of temporary birth control methods. The Ninth Plan (1997-2002) states the aim of bringing the awareness of “Small Family for Happy Family” to the rural populace and to reduce population growth through improvement in the health of women and children.

The government’s pro- family planning and small family position is also evident in the media and even in the school curriculum. Advertisements for family planning are found throughout the Kathmandu Valley on buses, taxis, and billboards, and on the radio, on television, and in newspapers. As early as grade six, school children begin to learn the

importance of family planning for the development of the country, and the difference

between permanent and temporary methods of contraception. By grade seven, children

147

learn more about various methods of contraception. In the textbook “Our Environment

and Population” for class seven, there is an illustration of a family planning camp bearing

the slogan “small family happy family.” The same textbook for class eight discusses the

“culture of low fertility:” how industrially advanced countries, the rich countries of

Europe, North America, and Japan, prefer small families. “They are happy and satisfied with one or two children. ‘A new baby or a new car’ is their motto. So often they decide

in favour of the car.” Above and beyond encounters with family planning messages

through school and advertising, the exposure to new ideas in general through an increase

in the prevalence of education, through encounters with global media (see Liecthy 2003),

and through a general trend in the consumption of modernity in the terms of development

created a favorable context for the acceptability of modern methods of contraception.

Despite these positive conditions for fertility decline, high infant mortality and low returns on investment in education continue to trouble people in Nepal. This combination of positive and negative conditions for fertility decline creates similar conflicting needs on the ground; people desire fewer children, but are still concerned about the reality of infant mortality and the need to have a son (and subsequently a daughter-in-law) to carry on the family.

The following account, told by an elderly but active mother-in-law about her eldest son’s wife who left the family out of shame, highlights potential dangers to having

few or no sons given the current structure of society and demographic realities. One of

the elderly woman’s junior daughters-in-law began by saying, “During our time, we

didn't know about this ‘happy family’ (sukhi pariwar).” The mother-in-law continued:

Before [people] didn't know, but now after having two or three children they do family [planning]; although this doing family planning, it doesn't suit some

148

people. It suits some, it doesn't suit others. Our eldest daughter-in-law – her one daughter is still living – she had two sons and then did family planning. The two sons died. And the mother became naked, no? It’s like that, it doesn't suit some. …Her daughter ran off, and her sons died. And didn't the mother become empty?! It’s like that. Now, I don't allow the daughters-in-law to do family planning. However many children they have, they will have. …After having them, may they raise them; after they get older, may they earn and eat.

As this story attests, a woman may be left metaphorically “naked,” or exposed and vulnerable, and “empty” in society if she produces no sons or loses the ones that were supposed to care for her.

Like this older woman, others were familiar with the “happy family,” two-child model that was commonly promoted on radio and television. Beginning with their establishment in Nepal in the 1960s, internationally funded family planning organizations spread the message that a small family was a happy family (a slogan that can be heard in different languages throughout the developing world). And indeed, women talked about two, otherwise three, children being enough. This model would not have been deemed acceptable locally, though, without the creation of a favorable environment by economic and social circumstances: the increased value of schooling, wage paying jobs, and modern accoutrements. Yet there is a problem with the “happy family” representation, a serious inherent contradiction that is apparent in the very logo of the Family Planning

Association of Nepal: the image of two parents with two children, one daughter and one

son. The presence of one son is crucial in this cultural context. How can a family ensure

such a perfect composition of offspring as in the logo? The same pragmatism that led

Nepali people to have fewer children in response to social and demographic

circumstances, when turned and applied to the need for sons, results in having children

until one has a son.

149

Son preference and fertility decline

The relationship between son preference and fertility decline operates in two

directions. First, son preference can create a lag in the intermediate stage of fertility

decline. Despite a general desire to have fewer children, families are driven to keep

procreating until they produce a son. Sex composition of the children, in particular the

presence of a male child, takes precedence over some general perception of having

“enough” children. For example, a study in Punjab observed that while educated women

desired smaller families, their desired number of sons fell less dramatically compared to

their desired number of daughters (Das Gupta 1987). Other evidence from India and

Korea is used to take this observation a step further, demonstrating that sex preference

slows demographic transition (Arnold, Choe, & Roy 1998; Larsen, Chung, and Das

Gupta 1998). Arnold’s study indicates that through the postponement of the end of one’s

reproductive career until the birth of a son, sex preference may exert its strongest effect

on fertility during the intermediate stage of demographic transition, slowing fertility

decline.

Second, fertility decline can create an intensification of sex bias. In a cultural context where sons are highly valued, a decrease in the number of offspring creates more pressure on families to produce sons in fewer tries. Das Gupta and Bhat proposed that as fertility decreases, sex discrimination could become more evident at lower parity births if the decrease in family size is not accompanied by a decrease in the desired number of sons (1997). This “intensification effect” is based on results from China and South Korea

that indicated that excess mortality of girls became more pronounced as fertility fell because parents discriminated more at each parity (Hull 1990; Zeng, Ping, Baochang, Yi,

150

Bohua, & YongPing 1993). Pressure to produce sons can result in families utilizing

medical technologies to manipulate the outcome of sex composition through sex-selective

abortion (Arnold, Kishor, & Roy 2002) or infanticide, but Das Gupta and Bhat caution

against overestimating the role of sex-selective abortion technologies in imbalanced sex

ratios, particularly in India (1997). Alternatively, pressure on women to produce sons also can lead to the reevaluation of the need for sons.

In Nepal both of these elements in the relationship between fertility decline and son preference are evident, but there has been no evidence that the pressure to produce

sons in fewer attempts has resulted in excessive mortality of girls or the use of sex-

selective abortion. Instead, a preference for sons makes some Nepalese willing to have

families larger than their stated ideal (Stash, 1996). A national level study, based on data

from the 1996 Nepal DHS, estimated the sex ratio at last birth for women who claimed to have completed their families or been sterilized at 146 males to 100 females (Leone

2003). This imbalanced ratio of boys to girls at last birth suggests that stopping behavior was driven by son preference. In Vishnumati case study discussions of the number of children that were necessary, the presence or absence of sons in the formula always changed the end sum. Many women, including mothers-in-law, said about their daughters-in-law, “This generation gives birth to up to three, no? Why have more than that?” But when asked if those three were daughters, a common reply was, “If they are all daughters? One son is needed. One son is needed for the funeral rites.” If the formula was changed to consist of one or two sons, then women said a daughter was needed. But while there were several examples of women having eight, even ten

151

daughters while trying to have a son, families abandoned their tries for a daughter soon

after reaching their desired family size.

Despite being positioned in the recent trends of comparatively lower fertility,

women were still expected to produce sons. In practice, this meant that the stakes were

raised for each birth. Nepali women thus found themselves located at the point of conflict between the ideals of lowered fertility and the value of sons. The lower the parity at which a son is born, the more quickly the family could breathe a sigh of relief.

But what were the reasons given for why sons are so important, and how susceptible are

they to change?

What good are sons? (the older generation of mothers-in-law)

Women debated the need for sons; some debated internally, some debated it with

other women in their household in my presence, and some thought that there was no

debate – of course a family needs to produce sons. Some women proclaimed defiantly,

“What good are sons?!”, but most conceded that they were good. And young women who had sons were relieved to be exempt from the discussion having any implications for them.

Women of the older generation, meaning those who were mothers-in-law, emphasized the importance of sons in two ways: in conducting funeral rites and in bringing a daughter-in-law into the house. Everyone agreed that a son was needed during the funeral rites to light his parents’ pyres. This is an important and pervasive Hindu belief, but people had difficulty explaining the significance behind the tradition. One

Brahman woman, the wife of a respected local Hindu priest, was in a position to explain:

152

We need one son because without a son we cannot do anything when the parents die. A daughter cannot do funeral rites of the parents according to our religion. A son has to do funeral rites… Whatever no-good, drunkard, thief and gambler a son may be, society cannot do funeral rites without a son. Even at the time of death you need a son because daughters are married off to other houses; they will not be able to come home at the time of emergency. From that point of view, too, a son is badly needed. Otherwise there won’t be a single person to give you a drop of water on your death bed. Who will take care of the parents in the family if a son is not there?

Younger women declared that they did not understand these “traditions.” If families do not have a son, they may find a nephew to do the funeral rights; according to older women a daughter just will not do. But like with most ideals, situations arise that are far from perfect: situations in which there is no son, or nephew for that matter. Anecdotally I heard of brothers and even daughters lighting the pyre.

Another high-caste mother-in-law summed up the need for sons in terms of the role of daughters. She joked loudly with the mother-in-law and two daughters-in-law from the house next door, “The daughter runs off – if she is given away she goes, if she isn’t given away she goes. If there is a son, he will care for [the parents].” She was referring to the fact that whether a woman elopes or her marriage is arranged, eventually she leaves her parents’ household. Everyone recognized that fact. Sons, on the other hand, remain with the family and care for the parents as they age. The Jethi Buhaari

(daughter-in-law married to the eldest son) of the family added:

After one becomes old, parents, we old people, won't be able to work. Can’t do field work, can’t care for livestock. For that a son is needed. After a son is born, a daughter-in-law (buhaari) will come. After the buhaari comes, they will manage things and do the fieldwork. They will have babies. And for these things isn't a family necessary? Without a son, where does a buhaari come from?!

Sons are needed for a much more practical and pressing reason than conducting funeral rites to a married woman and her family – a son brings a daughter-in-law into the house.

153

More than an issue of elders telling one how to behave according to religion and

tradition, this is a matter of daily responsibilities and the functioning of the household

and family. The introduction of a daughter-in-law into a family through marriage significantly changes the allocation of responsibilities in the household: the matriarch and

daughters in the family pass duties such as fetching firewood and fodder, cooking,

washing clothes, and cleaning on to the new wife. Thus the new wife bears the greatest

burden of time and labor intensive domestic duties and has the lowest status in the

hierarchy of her new family (Bennett 1982; Cameron 1998; Jeffery, Jeffery, & Lyon

1988). Over time, her position improves with the birth of children, especially sons, and

the introduction of another young wife into the family by her husband’s brother or

eventually by her own son (Das Gupta 1995).

Deliberating over giving birth again (the current generation of young mothers)

In Vishnumati, the pressure on individual women to give birth to sons came from a multitude of sources. The value of sons was reinforced subtly through popular culture, such as television serials, and in less subtle ways, such as doctors discouraging sterilization for women with only one son. But women experienced the pressure to produce sons most directly through their in-laws, their husbands, sometimes their neighbors, and, notably, themselves. Here I focus on young mothers who had no sons at the time of my research, for they were the ones who could be caught in the act of negotiating the pressure to give birth again. Women who had already given birth to sons tended to downplay the issue. All of the young mothers in my case studies who did not yet have sons were perplexed over whether or not they would give birth again.

154

The type of household that women had been living in for several years was related

to how bold they were in stating their reproductive preferences. Their attitudes varied along the lines of household structure, with women living in nuclear families being more vocal and adamant about not needing a son than women in joint families. In multigenerational joint families, a new wife assumes the lowest position in the hierarchy of family members. She works the most, eats the least of choice foods, and has little to no say in household decisions. Family members live in close proximity with little privacy. It is no surprise that when asked their major source of happiness, women in joint families said “when the family members get along well with one another.”

Nuclear family households have always composed a sizeable percentage of overall households in the joint family system found among Hindus in Nepal and northern

India. When the number of members in the stem family outgrew its ability to support them, or the household ran out of space, sons would break off and form a new household.

Another common reason for a split, evidenced by the countless films and television shows that portray it, is discord between the mother-in-law and a daughter-in-law. Other common reasons for forming a nuclear household in my survey were migration in search of economic opportunities or in flight of violence caused by the Maoist insurgency, or eloping in a love marriage. Thus at any given point in time (in a fertility regime in which multiple sons were the norm) one could find more nuclear families than joint families, but without negating the joint family as the overarching system and ideal.

In a joint family system, young couples are absorbed into the larger economic unit of the family and provided a cushion that can help reduce the costs of childbearing. In

Vishnumati, living in a nuclear family was often characterized by being in a state of

155

economic hardship. Assuming that the split from the stem is an amicable one, a son and

his wife may claim some of his inheritance but will be unable to depend upon the stem for any additional major financial support, and most of the son’s inheritance is usually

tied up in land and therefore inaccessible. One common reason for splitting is to search

for a better source of income in a more urban area. If the separation results from a fight,

then undoubtedly the resulting nuclear family will be in a state of financial difficulty

because a dependable, sufficient cash income is difficult to come by. Therefore rather

than splitting from the stem family, sons often rely upon the wealth or agricultural fruits of their parents in an economy that cannot support enough wage-paying jobs.

As a result of the financial hardship, women in nuclear families are more likely to be working either in a small one-room local shop nearby, in neighbors’ fields, or if low- caste, as tailors. In addition to working outside of the home, women in nuclear families have more autonomy within the home. They are responsible for running the household, making decisions regarding food, and, if working, they have some say in the household finances. The key component that allows all of this to happen is the lack of traditional familial hierarchy within the home. Instead of being the lowest person in a multi- generational household, the woman is second only to her husband.

High-caste, joint-family cases

Women in joint families contextualized the decision to give birth again within their larger family unit. They considered the desires of various family members, especially parents-in-law, and how not giving birth again would disrupt family harmony.

However the women did not present themselves in opposition to the other members, at

156

least in regard to reproduction. Instead they had internalized the importance of producing a son, and they presented themselves as being in a state of internal conflict and indecision

over whether to try again. They didn’t want to give birth again, yet they felt it was their

duty to produce a son.

Durga, a young high-caste woman with two daughters, lived in a sizeable cement

home with her many joint family members. Overall she expressed much suffering in her

husband’s household. She characterized her marriage in this way: “Ours is an arranged

marriage, so we don’t know what love is…” But after she gave birth to a child, she said

her husband loved her a little more. “It is said that a woman will be respected and loved

only after she has gotten a baby.” My research assistant then joked, “Due to the feeling of not being able to elope with another man?” The young woman laughed and agreed, and continued along that line saying, “After having many children our responsibilities, too, will increase and (there will be) no ability to go anywhere.” After she gave birth, she

began to wear kurtha salwar (the long tunic and loose pants worn by women throughout

South Asia) around the house instead of sari. This signified a slight increase in her

freedom, since in villages like this one mostly unmarried women wear kurtha salwar and

married women wear sari. Having a baby was viewed as a curb on a woman’s potentially

threatening sexuality, and it reduced her desirability to other men; women are unlikely to

leave their marriages because no one would want to take in a woman with a child.

Despite how strict her parents-in-law were, particularly her mother-in-law

(saasu), she said that she was the person most disappointed when she gave birth to her

second daughter. Although some husbands will beat their wives if they do not produce

sons, she reported, her husband had not said anything to her. In fact, “When I used to get

157

angry with the daughters, with being unable to get a son, their daddy used to convince me

that there is no difference between son and daughter,” she said. “The second time when I

gave birth to a daughter, he scolded me not to get angry because I gave birth to a

daughter… I had asked the doctor whether I had gotten a son or daughter. If a daughter,

how to go back home?! When I asked like that he scolded me.”

Her husband and parents-in-law did not say anything to her about getting a

daughter for the second time. She had internalized the necessity for sons; and yet, at least

by her own admission, her husband’s family members (ghar) had not expressed their

disappointment that she had not yet produced a son. When I asked if she needed a son, at

first she said no. Later Durga admitted, “No, it would have been better if she [the second

daughter] had been a son. Thinking that I would get a son, I gave birth [the second time].

I didn’t get a son, and what can be done?” She drew attention to the fact that she could not control the sex of her children, but despite that she felt the disappointment of not getting a son. In the end she concluded, “If I have [another child], then [it will be] much later. After they [the two daughters] get bigger.” She was 27 at that time. Her daughters were seven and two years old. Everyone was implicitly expecting her to have another child, but she was reluctant.

Another young, high-caste woman also expressed the need for sons primarily in terms of her own internalization of the expectation. Sita had two daughters, at that time age four and seven, and when asked if she needed a son, she said, “I don’t need one, but others disagree.” Others, she said, meant her parents-in-law. “They say they need at least one.” When asked about her husband’s preferences, she replied:

He doesn’t need. Now he says not to give birth. So my husband says, “Why give many births? It will be alright if we don’t give another birth.”

158

We are only two; we are not able to do [like that] because of others. I did temporary family planning. I haven’t adopted a permanent measure. Daughters also are grown up. [Pause…] I feel like I don't know what to do. It is uncomfortable. It would have been so nice if the youngest daughter had been a son. It would have been carefree (ananda). No one would have been able to say anything. In our society that isn't good. It is said that a son is needed. And also Deuraani (her husband’s younger brother’s wife) had a son, I had two daughters. When sisters and aunties came they told me that they wanted to see one from me. Everyone that came [said that]! I couldn’t do anything. What to do.

When asked what her husband thought after the birth of the second daughter, she assured

us, “He didn't behave badly, he didn't. But I felt bad when I gave birth to a daughter in

the hospital at that time… myself.” Contradicting her earlier general comment that her

parents-in-law needed a son, Sita said that her parents-in-law were supportive after the

birth of her second daughter. Her parents-in-law told her that a daughter is just like a son

and not to feel disappointment (dukha). But this statement was a consolation, not a negation of the need for a son. Even though no one in her ghar said anything negative to

her about giving birth to a second daughter, she had internalized the expectation of a son.

“I felt like, ‘What will others say because of having another daughter?’ So I felt bad. But

no one said anything. I felt bad myself.” Despite the fact that no one said anything, Sita

was not satisfied (literally her “heart did not understand”). Like with many expectations

in Nepali families, there was no need to express aloud that which was already understood.

Towards the end of the interview, an hour into my fourth discussion with her over

a period of seven months, she contradicted her earlier statement that she did not need a

son:

I will see once. I will give birth once more. And if it [a son] doesn’t happen, what to do, no? …They told me to give birth. No one obeys me. What to do? …I will try one more time. If again it is a daughter, then I will go to my maiti (woman’s parental home), get [the sterilization] operation there, and return.

159

Whether or not my husband marries another. It doesn’t matter to me. I am thinking that now.

Sita feared that her husband’s “heart might change” and he might take another wife. This

concern was commonly expressed by women when they talked about acting against the

will of their husbands. Again she repeated,

But I have fear that because of having daughters, my husband may marry (a second woman) later. Will he do or not. That is in my heart/on my mind (mero manmaa chha). I gave birth to daughters. If I do not obey what others say, I have fear of that in my heart. There will be no peace of mind.

“Because of everyone complaining, I plan to see once… If it is not in my karma, I will

not have (a son); what to do then, no?” She asserted that her third birth, in hope of a son,

would be the last – even if it did not produce a son. This was a change from women of

the previous generation, who told stories of a few women giving birth seven, even ten,

times in the pursuit of a son.

She felt that she should not adopt permanent family planning because of the

potential problems it would cause with her in-laws and her husband, even though at one

point she said her husband supported her in not giving birth again. She and her husband

lived with his parents and his younger brother, his wife, and their newborn son. The

couple had a comfortable life, although their substantial home was gradually becoming

smaller with the birth of additional children. After giving birth, she was allowed to stay

at her maternal home for two full months and take a full year’s rest from work other than

cooking and cleaning. Her parents-in-law were kind, and they did not mistreat her like some. In weighing the outcomes of giving birth again or not, it seemed that she had more

to lose – the care and security provided by her husband and parents-in-law, and the comforts of their home – by not giving birth.

160

This couple may have agreed on some idealistic level that they do not need a son,

but on a more realistic level they did need one. Much like the shift in her attitude towards giving birth during the interview itself, I expect that in the end she will try for a son one more time even though for the past 5 years of her life she has used Norplant to avoid becoming pregnant. She and her husband had postponed the decision of whether to try once more for a son. Sita seemed to be testing the limits to the rules about producing

a son, waiting to see what would happen and how bad the situation would get, before she

allowed the possibility of becoming pregnant.

A description of Sita’s story begs the question of where are “the rules” regarding

producing sons – where do they reside? Who enforces the social norm of having sons,

and where does the pressure come from in this case? Her husband and parents-in-law

didn’t say anything to her directly, but other women who visited after Sita’s sister-in-law

had a son all asked the same question of when she was going to get a son. Sita referred to

her fear of society (samaaj), and said that “they” told her to try again. The expectation

was implicit; it was something that everyone understood. Where does the pressure to

have sons originate, and how it is enforced?

When women said that they did not need a son, they were effectively arguing that

they did not. They were arguing against something, against the expectation for women to

produce sons, rather than merely stating a preference. And I propose that, in addition,

they were arguing between these two conflicting ideals: a small family and a family with

sons. In the end of the interview described previously, Sita concluded,

So, here’s my problem. People don’t understand. They say that [a son] is needed at the time of death. They say that “the door will be closed.” So “the door has to be opened.” I can't understand people’s talk! “The door will be closed at the time of death if we have only daughters,” they say. And if there is a son, whatever

161

kind of son, the door can be opened, it is said. What kind of thought is this of society?! Even if I quit thinking bad things, others will talk badly. Even if one tries to be satisfied oneself, others say bad things. What can you do?! Society is like that.

Negative “talk,” then, is a thing to be feared and avoided. Women spent time thinking,

worrying, and discussing what other people might say. A common way of phrasing

responses to what sounded like sometimes obvious, sometimes strange anthropological

questions was, “It is said…” (bhanchha). It is said that women should fast on such a date; women should stay quiet and underneath their husbands, it is said.1 This pointed to

an understanding of some consensus on how women should act.

The pressure to have sons comes from people’s perceptions of how other people

view, and talk about, their behavior. The forms of social structure currently in place also

reinforce this pressure: the institutions of marriage, family, and dharma, and the place of

women in each of them. And women themselves are often the vehicles for this

enforcement, in terms of what family members or villagers say to one another, in terms of

the gossip that is assumed to occur, and in terms of some authoritative group of people

that is mysteriously behind all the passive voice “It is said” statements. Women like Sita

– who no longer believed in the unassailability of traditions such as the need for a son to

ensure one’s passage to the afterlife – still felt pressure to conform in order to avoid the

scorn and gossip of others.

Low-caste, nuclear-family cases

A few low-caste, low-class women who had only daughters were more defiant

about not giving birth again. These women were living in nuclear family households.

162

They claimed that their daughters were just as good as sons. Radha, who was 24 at the

time, lived in one rented room with her husband and ten-year-old daughter. She claimed

that living in a nuclear family situation did not relieve the pressure on her. She joked,

“My own husband has treated me just like the parents-in-law would have!” “If my husband had better behavior, I might want to have another. But now I don’t want to.”

She and her husband fought from time to time after he had been drinking, and he beat her severely during these occasions. Her husband squandered away much of his income outside the home, in local stalls that served food and alcohol. Like many women, she worried about him flirting with and bringing home other women. He had pursued other women in the past, even married women. “Even if he has not brought one home up until now, there is no guarantee that he will stay quiet,” she said.

Yet Radha did not hesitate in responding to the question of whether she was interested in having another child:

If he needs a son then I will ask him to bring another wife. After all, can we give birth to a son if we want one, or a daughter if we want to have a daughter? If we get a daughter, their behavior is so different; they insult. In the Nepali custom if we don't have a son we are not given food to eat also. They mistreat us if we give birth to a daughter. There is a Nepali proverb: if give birth to a son, then feed with mutton meat (khashi); and if a daughter then feed with pumpkin (pharsi). The very behavior of the family (ghar) changes with the birth of either son or daughter. So how can we have that? We are not machines. So what guarantee is there that my next child would be a son? Suppose I get only a series of daughters, then what?! It will be very difficult to care for all the girls. We can't kill them, we have to look after them well, we have to earn. It is very difficult.

Radha’s words summarized the opinions of several young mothers living in nuclear

families when she said, “Look at my own husband. He neither serves his parents in times

of need nor cares for his wife. So I don't feel that a son is highly needed. Instead, I feel

1 Linguistically this “it is said” also can be subverted easily by using a certain tone of voice, in the sense of, “That’s what is said, but I think it’s ridiculous.”

163 that the daughter will give more love.” Another woman living in a nuclear family put it this way, “A son is no longer needed. What have sons done in recent times?! …If I can do well for my daughter, she too will take care of me. In the case of earning well, she too can do well.” But in response to her husband’s opinion that girls are worth little and do not need to be educated, Radha commented, “He is also correct in some ways. Because a daughter is married off, whereas the son will be compelled to look after the old parents.”

Much like in the previous cases, as this discussion progressed she moved from speaking in an idealistic way to admitting the reality of the situation for her and for other women who cannot leave their husbands. Women believed that they were constrained to marriages for economic reasons and the sheer fact of having nowhere to go. Living alone as a single woman invited accusations of prostitution, and returning to the home of her birth invited scorn from the community and members of her family who resented the extra burden she and her children placed on their income.

Even the young women who were the boldest in their opinions about not needing a son could not overcome the social fact that daughters leave their parents at the time of marriage. The patrilocal, joint family model of living undermined the logic of the women who said that their daughters were just as good as sons. Women living in nuclear families hoped that their sons would stay with them and bring daughters-in-law into the home – transforming the household back into a joint family. So even though young women were dismissive of traditions related to their culture and religion such as needing a son for funeral rites, they couldn’t escape their underlying assumptions about patterns of residence and marriage.

164

A careful reading of women’s comments reveals as much accommodation as

opposition to the pressures that they faced to reproduce in socially prescribed ways.

Although I highlighted their discursive and bodily oppositions to the need for sons, I

predict that in the end these women’s oppositions will be limited to merely the

postponement of giving birth again. Their agentive status, however, is evident in their

indecision and conflicted feelings over giving birth, and acted out through this

postponement – regardless of whether the postponement was a conscious effort to test the

limits or just the avoidance of an unwelcome task.

Despite women’s articulated experiences of a range of suffering related to their

families and society, ultimately women did not work toward subverting the system. The fact that women identified the sources of their subjugation suggests that the ways power

operates in society had not become so naturalized that they were not aware of it. Sita’s

vacillation over trying one more time for a son indicated her awareness of her own

desires as in opposition to those around her, and her agentive status. For her, too much

was at stake for an outright refusal to give birth again – her comfortable life, the affection of her husband, and harmonious relations within the household.

Not having any viable alternatives also held women in check. Radha remained in her marriage because she foresaw much greater problems if she tried to escape it. It was easier, more effective, and more economically and socially rewarding to strategize within

the rules than to subvert them. The literature from other parts of South Asia suggest that it is only a matter of time before examples of alternative options for average women begin to appear through pioneering and creative cases. In the meantime, women

attempted to maximize their positions by working within the social rules and shifting sets

165

of discourses that govern reproduction in the ways described here, even while wishing

that things were different.

These cases also suggest that while classic cultural and religious explanations for needing a son are fading, the multi-generational family ideal remains strong. Having (or expecting to have) a joint family household influences a couple’s desire for sons because, at least in part, of a need for a daughter-in-law to replace the daughters lost to other households through marriage. Despite their resistance to the need for sons, women are unable to escape the logic that 1) increases in the potential contributions of women to the household benefits them only through daughters-in-law, and 2) having a son results in the household gaining a daughter-in-law. Having (or merely expecting to have) an extended family household influences a couple’s desire for sons. Women’s verbal opposition to a higher value placed on sons in their culture ultimately does not overcome their desire for a daughter-in-law who will provide labor and care, and ensure the continuation of the ghar (home/family). The patrilocal extended family system thus continues to drive observed son “preference,” despite changes in gender norms and education.

In the future, if parents begin to see that their investment in girls pays off through

remittances from their daughters’ employment, the counter-discourse to the need for sons

will grow – but mainly in nuclear family settings. This possibility was foreshadowed by

the women in nuclear families who proclaimed, “Can’t my daughter care for me?”

Without a significant change in the patrilocal and joint family system, though, daughters

cannot care for their parents. Once they are married, daughters must serve their parents-

in-law, either directly through their labor in the household or fields or their income from

employment outside of the home.

166

Alternatively, if what appears to be a temporary nuclear stage in a joint family system turns out to be a building momentum toward a nuclear family system, at the household level daughters would not need to be replaced by daughters-in-law. Either son or daughter could provide and care for their parents, or the aging parents might be placed in a growing number of facilities for the elderly. Like many trends that appear first in

India and later in Nepal, the development of the “bad family” (Cohen 1998; Lamb 2000) may take hold in Nepal. In any of these alternative scenarios to the patrilocal joint family, the patriarchal nature of society would not necessarily subside, but the findings of this research indicate that in these scenarios son preference would weaken.

CONCLUSION

REPRODUCTIVE REALITIES IN NEPAL

In this project, I highlight the contradictions that captured women’s attention at a

given period in their lives, but also women’s position amidst what I as a social scientist

see as contradictory social forces: fertility decline and son preference. I present young

married women’s vacillations over the need to produce a son within the context of fertility decline and the discourses surrounding it, and in contrast to the some of the older

generation’s religious explanations for the need for sons. Although several young

mothers espoused that girls were just as good as boys, they could not overcome the

implications of the dominant joint family system for needing a son to bring a daughter-in-

law into the household. Because of this household structure, women and men were

concerned with producing sons and subsequently daughters-in-law who would assist with

household work, income, and elderly care.

On a more theoretical level, I argue that women’s conflicted feelings and

rationales regarding procreation provide a glimpse into the dialectical process of “culture

in the making.” I explain how women’s positions are affected by the interaction of social

hierarchies of caste, class, and gender, and how women’s positions also change over the

life course in respect to the family cycle. I describe how social relations affect women’s

health during pregnancy, birth, and the postpartum period using the language of

embodiment. In these ways I account for the structural constraints on women’s agency.

167

168

And finally, I offer the simple act of postponement of birth as a site to analyze agency, and I present the reproductive body as a locus of the dialectic of structure and agency. In closing, I submit a few final words on the topic of agency, culture, and women’s health.

Multiple types of behavior may at first glance appear to be single type because of their similar role in maintaining the status quo. The kind of behavior that appears on the surface as system reproduction may be one of (at least) two things: unconscious or conscious behavior according to the rules of the system. I identify two aspects of unconscious behavior that relate to this project: routinized behavior and a prevailing sense of fatalism. The former is a result of the hegemonic aspects of culture by which humans are enculturated into following certain mundane everyday habits, whereas the latter is a characteristic of certain cultural systems that devalue individual human agency, either of all people or certain groups, and place faith in specific authoritative figures either natural or supernatural.

Reproducing the system may be merely an act of routinized behavior, of not questioning one’s actions, motives, or constraints. Ortner points out that action does sometimes “proceed with little reflection” (1984:150) (especially in the domestic domain), and that despite a focus on action both Bourdieu and Giddens recognize “the central role of highly patterned and routinized behavior in systemic reproduction”

(1984:150). Not everything, Ortner reminds us, is pragmatic choice, active calculating, and strategizing. Across all cultures, a certain amount of daily behavior is seldom, if ever, questioned or reflected upon.

Fatalism at the individual level differs slightly from the description of routinized behavior; it is the product of a larger cultural sense of not having the ability to affect 169

outcomes. Depending on one’s perspective, it can be viewed as a sense of helplessness

or a calm resolve to face whatever may come. This can come about through belief in

someone or something else being in control of life, whether that is a general sense of fate, a deity or religious figure, or some other person(s) such as community leaders or the men

in a household. Any of these can cause people to doubt the efficacy of individual human action or of their actions as a particular individual. In many patriarchal societies for example, being a good woman can be tied up in exhibiting a strong sense of resolve in the face of suffering; thus it can be difficult to empower a woman in this context because

trying to change things looks like weakness, mere complaining, or an inability to cope

with hardship. Subjugation and isolation of certain groups may also result in a feeling of helplessness.

Alternatively, acting in a way that reproduces the system may actually be a conscious choice. It may result from a strategy in which an individual recognizes the constraints placed upon him or her and chooses to act within the current system, rather than challenging it, in order to maximize the benefits (or minimize the suffering, as would be the case with oppressed groups). Anthropologists have provided examples of expressions of resistance that did not result in a direct challenge to the system, but that, nevertheless, evidenced a sophisticated grasp of the situation and the oppression of groups/relations of power and the existence of alternative discourses (Raheja and Gold

1994; Abu-Lughod 1999; March 2002).

Most women in my study had little agency in regard to their reproductive health, particularly during pregnancy and birth. Several women told stories of life-threatening

complications during or after giving birth at home, and not being taken to the hospital 170

(despite the availability of an ambulance and the good fortune of a teaching hospital

being located only around 20 minutes away). They felt unable to control or influence such situations. However, most women in the study were self reflective if not strategic

about their reproduction, in terms of their use of modern contraception and their concerns

about bringing or maintaining harmony in the relationships that mattered to them most. It is thus “important to ask how people themselves conceive of their own actions” (Ahearn

2001:113) in order to discover how people attribute responsibility for events, and in what

realms people perceive themselves to have or not have agency and the meaning of those

distinctions. These distinctions plus the dynamic quality of people’s conceptions of

actions and agency demonstrate “the need for anthropologists to ask not only what

agency means for themselves as theorists, but what it means for the people with whom

they work, and how those meanings may shift over time” (Ahearn 2001:113).

This discrepancy in a lack of agency in matters concerning health and the

existence of agency in controlling the timing and ultimate number of births demonstrates

that there was not a uniform experience of agency across all procreation-related topics;

women felt that they had agency in some realms, but not in others. I attribute this in part

to a long history in Nepal of governmental and private agencies privileging messages and

programs aimed at fertility decline. I propose that women, and most likely couples, have

been empowered by the family planning propaganda to have smaller families. This

operated in tandem with the urbanization of villages in Nepal such as Vishnumati that are located in relatively close proximity to urban centers or along major highways in the middle hills and tarai regions. Such messages do not make sense locally in the remote rural areas of the country, hence the divide in total fertility rates between rural and urban 171 areas of Nepal. Moreover, the message that fewer children equal greater economic prosperity directly links limiting births with the desirable trappings of modernity. A globalized morality of the “small, happy family” was underlying and enabling women’s discussions of procreation, but their immediate concerns focused on negotiating their desires and those of the people who impacted their lives most directly. Women’s strategies may have operated on an intimate scale, but they were no less a reflection of larger structural forces at work. Therefore I employ a critical phenomenological approach: one that accounts for women’s embodied experiences of structural forces in their intimate relationships and even in their bodies, and concomitantly, women’s bodily assertions.

A more fully gendered account is needed in this regard, for although women mentioned how nowadays they must educate and provide well for their children, overall they emphasized maintaining good relations with their spouse and/or their spouse’s family above all else. It is possible that men would place more value in discussions of reproduction on personal financial issues and societal issues of economic development.

Future research should address the gendered interpretations and internalizations of family planning messages and the meaning of modernity.

In conclusion, women were not reproducing the system through routinized behavior. They recognized the constraints placed upon them as women in the context of the family and of the larger society, and they weighed the risks of challenging gender norms or the expectation to give birth again. One would imagine that the agency that young married women exhibited in regards to becoming pregnant could be transferred to gaining more influence over the conditions of birth itself – in recognizing that 172

complications during home births might arise, that trained birth attendants can identify

those complications earlier, and that emergency transportation to a hospital needs to be in

place. The practical application of this research, then, is to work toward that goal. If as

much money for creating services and public education messages were directed toward improving maternal health as historically has been aimed at lowering fertility in Nepal, this goal should be attainable.

Overall this is not a story of resistance, and hardly one of active protest. Social

change is sometimes caused by such easily discernable actions, but more often it is a

subtle, ongoing process. This account is an exploration of the latter. It is a mixture of

both quiet and defiant voices. Some women voiced concern, ambivalence, and

indecision, while others voiced clear frustration with the cultural and familial

expectations placed upon them as women. Through these voices, by presenting their

variation in response and yet similarity in subject of concern, I have attempted to perform

a “…careful accounting of the processes by which accommodation and opposition to

dominant discourses are intertwined” (Ginsburg and Rapp 1995:11). These glimpses into

the minutiae of women’s reproductive lives and decision making provide insight into

social change of the more everyday variety, and captures – for a moment – the dynamic

flow of the dialectic of culture.

Women were reproducing hierarchy by working primarily within the current sets

of gender norms and not against them and through their intentions to continue the joint

family tradition. However they were not reproducing a clone of the social hierarchies of

their mothers-in-law’s generation, but an altered version of hierarchies. This new version of hierarchy ultimately will be defined by the women’s children. The young mothers in 173

this study were truly embodying social change, for they were the generation caught

amidst a sudden onslaught of new ideas, technologies, and discourses. They were the

generation located between parents-in-law who valued hard work and obedience and

children who communicated via the internet and imitated styles of western hip-hop

clothing in their dress. Young mothers’ ideas about the world had been influenced by their own formal education, in contrast to the typical lack of schooling exhibited by women the generation above them, but most did not think that they were in a position to actualize those new ways of thinking. Thus the ways they actively defined culture – their acts of agency – showed up in mostly subtle ways such as postponing giving birth or debating the need for sons. Their children, though, will be in a position to live out the ideals of their mothers more fully. The women’s openness to change combined with a new set of historical circumstances and social pressures for their children once they reach adulthood will allow an unprecedented freedom for action in Nepal (outside of unique cities like Kathmandu). The young children whom I met during my research will read about their mother’s situations with amazement, if they even desire to hear about what will seem to them as such distant and traditional stories.

BIBLIOGRAPHY

Abu-Lughod, Lila 1986 Veiled Sentiments: Honor and Poetry in a Beduin Society. Berkeley: University of California Press. Ahearn, Laura 2001 Invitations to Love: Literacy, Love Letters, and Social Change in Nepal. Ann Arbor: University of Michigan Press. — 2001 Language and Agency. Annual Review of Anthropology 30:109-137. Arnold, Fred, Minja Kim Choe, and T.K. Roy 1998 Son Preference, the Family-Building Process and Child Mortality in India. Population Studies 52:301-315. Arnold, Fred, Sunita Kishor, and T.K. Roy 2002 Sex-Selective Abortions in India. Population and Development Review 28(4):759-785. Atkinson, Jane Monnig, and Shelly Errington, eds. 1990 Power and Difference: Gender in Island Southeast Asia. Stanford: Stanford University Press. Axinn, William G. 1992 Family organization and fertility limitation in Nepal. Demography 29(4):503-521. Axinn, William G., Thomas E. Fricke, and Arland Thornton 1991 The microdemographic community-study approach: Improving survey data by integrating the ethnographic method. Sociological Methods & Research 20(2):187-217. Axinn, William G., and Scott T. Yabiku 2001 Social change, the social organization of families, and fertility limitation. American Journal of Sociology 106(5):1219-1261. Bennett, Lynn 1982 Dangerous wives and sacred sisters: social and symbolic roles of high- caste women in Nepal. New York: Columbia University Press. Berreman, Gerald D. 1971 The Brahmannical view of caste. Contributions to Indian Sociology 5. Beteille, Andre 1965 Caste, class and power: Changing patterns of stratification in a Tanjore Village. Berkeley: University of California Press. — 1996 Caste in Contemporary India. In Caste Today. C.J. Fuller, ed. Delhi: Oxford University Press.

174

175

Bougle, Celestin 1971 Essays on the caste system. Cambridge: Cambridge University Press. Bourdieu, Pierre 1977 Outline of a theory of practice. Cambridge: Cambridge University Press. — 1990 The logic of practice. Stanford, California: Stanford University Press. Bourdieu, Pierre, and Loic J. D. Wacquant 1992 An Invitation to Reflexive Sociology. Chicago: University of Chicago Press. Bradley, Candice 1995 Women's empowerment and fertility decline in western Kenya. In Situating fertility: Anthropology and demographic inquiry. S. Greenhalgh, ed. Pp. 157-178. Cambridge: Cambridge University Press. Caldwell, John C. 1982 The wealth flows theory of fertility decline. In Theory of fertility decline. J. Caldwell, ed. Pp. 333-351. New York: Academic Press. — 1998 The Global Fertility Transition and Nepal. Contributions to Nepalese Studies 25:1-7. Cameron, Mary M. 1998 On the edge of the auspicious: Gender and caste in Nepal. Chicago: University of Illinois Press. Carter, Anthony T. 1995 Agency and fertility: For an ethnography of practice. In Situating fertility: Anthropology and demographic inquiry. S. Greenhalgh, ed. Pp. 55-85. Cambridge: Cambridge University Press. Central Bureau of, Statistics 2002 Population of Nepal: Village Development Committees / Municipalities Population Census 2001. Clifford, James 1986 Introduction: Parial Truths. In Writing Culture: The poetics and politics of ethnography. J. Clifford and G. Marcus, eds. Pp. 1-26. Berkeley: University of California Press. Cohen, Lawrence 1998 No aging in India Alzheimer's, the bad family, and other modern things. Berkeley: University of California Press. Collier, Jane, and Sylvia Yanagisako, eds. 1987 Gender and Kinship: Essays toward a Unified Analysis. Stanford: Stanford University Press. Csordas, Thomas J. 1990 Embodiment as a Paradigm for Anthropology. Ethos 18(1):5-47. — 1994 Introduction: the body as representation and being-in-the-world. In Embodiment and Experience: The Existential Ground of Culture and Self. T.J. Csordas, ed. New York: Cambridge University Press. — 176

1999 The Body's Career in Anthropology. In Anthropological Theory Today. H.L. Moore, ed. Pp. 172-205. Boston: Polity. Das Gupta, Monica 1987 Selective Discrimination Against Female Children in Rural Punjab, India. Population and Development Review 13(1):77-100. — 1995 Lifecourse perspectives on women's autonomy and health outcomes. American Anthropologist 97(3):481-491. — 1997 Kinship systems and demographic regimes. In Anthropological demography. D.I. Kertzer and T. Fricke, eds. Pp. 36-52. Chicago: University of Chicago Press. Das Gupta, Monica, and P. N. Mari Bhat 1997 Fertility decline and increased manifestation of sex bias in India. Population Studies 51(3):307-315. Das Gupta, Monica, et al. 2003 Why is Son Preference so Persistent in East and South Asia? A Cross- Country Study of China, India and the Republic of Korea. The Journal of Development Studies 40(2):153-187. Das, Veena 1977 Structure and cognition: aspects of Hindu caste and ritual. Delhi: Oxford University Press. — 1992 Reflections on the social construction of adulthood. In Identity and adulthood. S. Kakar, ed. Pp. 89-104. Delhi: Oxford University Press. Derrida, Jacques 1976 Of Grammatology. G.C. Spivak, transl. Baltimore: Johns Hopkins University Press. di Leonardo, Micaela 1991 Introduction: Gender, Culture, and Political Economy: Feminist Anthropology in Historical Perspective. In Gender at the Crossroads of Knowledge. M. di Leonardo, ed. Berkeley: University of California Press. Dirks, Nicholas B. 1987 The hollow crown: Ethnohistory of an Indian kingdom. Cambridge, New York: Cambridge University Press. Dumont, Louis 1966 Homo Hierarchicus. Chicago: University of Chicago Press. Dyson, Tim, and Mick Moore 1983 On kinship structure, female autonomy, and demographic behavior in India. Population and Development Review 9(1):35-60. Ehrlich, Paul 1978 The Population Bomb. New York: Ballantine. Errington, Shelly 1990 Recasting Sex, Gender, and Power: A Theoretical and Regional Overview. In Power and Difference: Gender in Island Southeast Asia. J.M. Atkinson and S. Errington, eds. Pp. 1-58. Stanford: Stanford University Press. 177

Fisher, William F. 2001 Fluid boundaries: Forming and transforming identity in Nepal. New York: Columbia University Press. Foucault, Michel 1966 The Order of Things: An Archaeology of Human Sciences. New York: Vintage. — 1978 The History of Sexuality: An Introduction. New York: Random House. Fruzzetti, Lina 1982 The gift of the virgin. New Brunswick: Rutgers University Press. Fruzzetti, Lina, and Akos Ostor 1976 Is there a structure to north Indian kinship terminology? Contributions to Indian Sociology n.s. 10(1):63-95. — 1976 Seed and earth: a cultual analysis of kinship in a Bengali town. Contributions to Indian Sociology n.s. 10(1):98-132. — 1984 Kinship and ritual in Bengal. New Delhi: South Asian Publishers Pvt. Ltd. — 1998 Hierarchy Revisited. In Changing patterns of family and kinship in South Asia. A. Parpola and S. Tenhunen, eds. Pp. 39-50, Vol. 84. Helsinki: Finnish Oriental Society. Fruzzetti, Lina, Akos Ostor, and Steve Barnett 1992 The cultural construction of the person in Bengal and Tamilnadu. In Concepts of person: Kinship, caste, and marriage in India. L. Fruzzetti, A. Ostor, and S. Barnett, eds. Pp. 8-30. Delhi: Oxford University Press. Fruzzetti, Lina, and Sirpa Tenhunen, eds. 2006 Culture, Power, and Agency: Gender in Indian Ethnography. Kolkata: Stree. Fuller, C. J. 1996 Caste today. Delhi; New York: Oxford University Press. Furer-Haimendorf, Christoph 1971 On the nature of caste in India: A review symposium on Louis Dumont's Homo Hierarchicus. Contributions to Indian Sociology 5. Geertz, Clifford 1973 The Interpretation of Cultures. New York: Basic Books. Gellner, David 1997 Ethnicity and Nationalism in the the World's Only Hindu State. In Nationalism and Ethnicity in a Hindu Kingdom: The Politics and Culture of Contemporary Nepal. D. Gellner, ed: Routledge. Giddens, Anthony 1979 Central problems in social theory: Action, structure, and contradiction in social analysis. Berkeley: University of California Press. Ginsburg, Faye D., and Rayna Rapp 178

1995 Introduction: Conceiving the new world order. In Conceiving the new world order: The global politics of reproduction. F.D. Ginsburg and R. Rapp, eds. Pp. 1-19. Berkeley, Los Angeles, London: University of California Press. Glasner, Barney G., and Anslem L. Strauss 1967 The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine Publishing Co. Greenhalgh, Susan 1990 Toward a Political Economy of Fertility: Anthropological Contributions. Population and Development Review 16(1):85-106. Greenhalgh, Susan 1995 Anthropology theorizes reproduction: Integrating practice, political economy, and feminist perspectives. In Situating fertility: Anthropology and demographic inquiry. S. Greenhalgh, ed. Pp. 3-28. Cambridge: Cambridge University Press. — 1996 The social construction of population science: An intellectual, institutional, and political history of twentieth-century demography. Comparative Studies in Society and History 38(1):26-66. Grosz, Elizabeth 1994 Volatile Bodies: Toward a Corporeal Feminism. Bloomington and Indianapolis: Indiana University Press. Guha, Ranjit, and Gayatri Chakravorty Spivak, eds. 1988 Selected subaltern studies. Oxford: Oxford University Press. Guneratne, Arjun 2002 Many tongues, one people: The making of Tharu Identity in Nepal. Ithaca: Cornell University Press. Gupta, Dipankar 2004 The certitudes of caste: When identity trumps hierarchy. In Caste in Question: Identity or Hierarchy? D. Gupta, ed. New Delhi: Sage. Haraway, Donna 1988 Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective. Feminist Studies 14:575-599. Harlan, Lindsey, and Paul B. Courtright 1995 From the margins of Hindu marriage: Essays on gender, religion, and culture. New York: Oxford University Press. Hartmann, Betsy 1995 Reproductive rights and wrongs: The global politics of population control. Boston: South End Press. Hofer, Andras 1979 The caste hierarchy and the state in Nepal: A study of the Muluki Ain of 1854. Innsbruck: Universitatsverlag Wagner. Holmberg, David H. 1989 Order in paradox: Myth, ritual, and exchange among Nepal's Tamang. Ithaca: Cornell University Press. Hull, Terence 179

1990 Recent Trends in Sex Ratios at Birth in China. Population and Development Review 16:63-83. Inden, Ronald B. 1990 Imagining India. Indianapolis: Indiana Univesity Press. Jeffery, Patricia, Roger Jeffery, and Andrew Lyon 1988 Labour Pains and Labour Power. London and New Delhi: Zed Books Ltd. Justice, Judith 1986 Policy, Plans, and People: Foreign Aid and Health Development. Berkeley: University of California Press. Kertzer, David I. 1991 Household and gender in a life course perspective. In Women, Households and Change. E. Masini and S. Stratigos, eds. Tokyo: United Nations University Press. Kertzer, David I., and Dennis P. Hogan 1989 Family, political economy, and demographic change. Madison: University of Wisconsin Press. Khandelwal, Meena, Ann Grodzins Gold, and Sondra Hausner, eds. 2006 Women's Renunciation in South Asia: Nuns, Yoginis, Saints, and Singers. New York: Palgrave Macmillan. Knodel, John 1997 A case for nonanthropological qualitative methods for demographers. Population and Development Review 23(4):847-854. Kolendra, P. 1976 Seven types of hierarchy in Homo Hierarchicus. Journal of Asian Studies 35(4):581-596. Krause, Elizabeth 2005 Encounters with the "peasant": Memory work, masculinity, and low fertility in Italy. American Ethnologist 32(4):593-617. Lamb, Sarah 2000 White saris and sweet mangoes: Aging, gender, and body in North India. Berkeley: University of California Press. Larsen, Ulla, Chung Woojin, and Monica Das Gupta 1998 Fertility and son preference in Korea. Population Studies 52(3):317-325. Leone, Tiziana, Zoe Matthews, and Gianpiero Dalla Zuanna 2003 Impact and Determinants of Sex Preference in Nepal. International Family Planning Perspectives 29(2):69-75. Levine, Nancy E. 1987 Caste, state, and ethnic boundaries in Nepal. Journal of Asian Studies 46(1):71-88. Liechty, Mark 2003 Suitably Modern: Making Middle-Class Culture in a New Consumer Society. Princeton: Princeton University Press. Madan, T. N. 1971 On the nature of caste in India: A review symposium on Louis Dumont's Homo Hierarchicus. Contributions to Indian Sociology 5:1-13. Mandelbaum, David G. 180

1970 Society in India. Bombay: Popular Prakashan. March, Kathryn 2002 "If each comes halfway:" Meeting Tamang women in Nepal. Ithaca: Cornell University Press. Marcus, George, and Michael Fischer 1986 Anthropology as Cultural Critique: An Experimental Moment in the Human Sciences. Chicago: University of Chicago Press. Marriott, McKim 1990 Constructing an Indian Ethnosociology. In India through Hindu Categories. M. Marriot, ed. London: Sage Publications. Mason, Karen Oppenheim 1986 The status of women: Conceptual and methodological issues in demographic studies. Sociological Forum 1(2):284-299. — 1987 The impact of women's social position on fertility in developing countries. Sociological Forum 2(4):718-745. — 1995 Reproductive Rights and Wrongs: The Global Politics of Population Control (Revised Edition). Population & Development Review 21(4):885-887. Ministry of Health and Population, New ERA, and Macro International Inc. 2007 Nepal Demographic and Health Survey 2006. Kathmandu. Mohanty, Chandra Talpade, Ann Russo, and Lourdes Torres, eds. 1991 Third World Women and the Politics of Feminism. Indianapolis: Indiana University Press. Mullany, Britta C. 2006 Barriers to and attitudes towards promoting husbands' involvement in maternal health in Katmandu, Nepal. Social Science & Medicine 62(11):2798- 2809. Narayan, Uma 1997 Dislocating Culture. New York: Routledge. Nussbaum, Martha C. 2000 Women and Human Development. New York: Cambridge University Press. Ortner, Sherry 1974 Is female to male as nature is to culture? In Women, Culture, and Society. M. Rosaldo and L. Lamphere, eds. Pp. 67-87. Stanford: Stanford University Press. — 2006 Anthropology and Social Theory: Culture, Power, and the Acting Subject. Durham: Duke University. Ortner, Sherry B. 1984 Theory in Anthropology since the Sixties. Comparative Studies in Society and History 26(1):126-166. Ortner, Sherry B., and Harriet Whitehead 1981 Introduction: Accounting for Sexual Meanings. In Sexual Meanings: The Cultural Construction of Gender and Sexuality. S.B. Ortner and H. Whitehead, eds. Cambridge: Cambridge University Press. 181

Pigg, Stacy Leigh 1996 The Credible and the Credulous: The Question of "Villagers' Beliefs" in Nepal. Cultural Anthropology 11(2):160-201. Quigley, Declan 1993 The Interpretation of Caste. New York: Oxford University Press. Quinn, Naomi 1977 Anthropological Studies on Women's Status. Annual Review of Anthropology 6:181-225. Rabinow, Paul 1977 Reflections on fieldwork in Morocco. Berkeley: University of California Press. Raheja, Gloria Goodwin, and Ann Grodzins Gold 1994 Listen to the heron's words: Reimagining gender and kinship in North India. Berkeley, Los Angeles: University of California Press. Riley, Nancy E. 1999 Challenging Demography: Contributions from Feminist Theory. Sociological Forum 14(3):369-397. Rosaldo, Michelle 1974 Woman, culture, and society: A theoretical overview. In Women, Culture, and Society. M. Rosaldo and L. Lamphere, eds. Pp. 17-42. Stanford: Stanford University Press. — 1980 The use and abuse of anthropology: Reflections on cross-cultural understanding. Signs 5:389-417. Scheper-Hughes, Nancy, and Margaret Lock 1987 The mindful body: A prolegomenon to future work in medical anthropology. Medical Anthropology Quarterly 1(1):6-41. Schneider, David M. 1974 A critique of the study of kinship. Ann Arbor: University of Michigan Press. Sen, Amartya 1990 More Than 100 Million Women Are Missing. New York Review of Books 20:61-66. Sharma, Ursula 1999 Caste. Philadelphia: Open University Press. Skinner, William G. 1997 Family systems and demographic processes. In Anthropological demography. D.I. Kertzer and T. Fricke, eds. Pp. 53-95. Chicago: University of Chicago Press. Stash, Sharon 1996 Ideal-family-size and Sex-composition Preferences among Wives and Husbands in Nepal. Studies in Family Planning 27(2):107-118. Strauss, Anslem L., and Juliet M. Corbin 1990 Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park: Sage Publications. Thornton, Arland, and Thomas E. Fricke 182

1987 Social change and the family: Comparative perspectives form the West, China, and South Asia. Sociological Forum 2(4):746-749. United Nations Population Fund 2008 Advancing the goals of the ICPD and the millennium summit. Vatuk, Sylvia 1975 Gifts and affines in north India. Contributions to Indian Sociology 9(2):155-196. Weiss, Linda 1999 Single women in Nepal: Familial support, familial neglect. Journal of Comparative Family Studies 30(2):243-257. Whelpton, John 2005 A history of Nepal. Cambridge: Cambridge University Press. Williams, Raymond 1977 Marxism and Literature. New York: Oxford University Press. Yanagisako, Sylvia 1979 Family and Household: The Analysis of Domestic Groups. Annual Review of Anthropology 8:161-205. Zeng, Yi, et al. 1993 Causes and Implications of the Recent Increase in the Reported Sex Ratio at Birth in China. Population and Development Review 19:283-302.