Anthropological Perspectives on Global Issues in Midwifery
by Robbie Davis-Floyd, PhD

© 2000 Midwifery Today, Inc. All Rights Reserved.

[Editor's note: This article first appeared in Midwifery Today Issue 53, Summer 2000.]

A distressing cross-cultural trend is showing up in the growing body of anthropological literature about midwifery and birth in the developing world. From Tanzania to Papua New Guinea, anthropologists who observe professional midwives giving prenatal care and attending births increasingly note that, far from the midwifery ideal, professional midwives often treat women very badly during birth, ignoring their needs and requests, talking to them disrespectfully, ordering them around, and sometimes even yelling at them and slapping them. At the same time, and in direct correlation, the professional midwives are themselves often treated badly by the healthcare systems in which they work. They are almost always underpaid, are frequently mistreated by physicians who rank above them in the medical hierarchy, and generally work long hours under stressful conditions that often include inadequate facilities and equipment and too many women with too few midwives to care for them well. In short, professional midwives are often trapped in the biomedical healthcare system, a system that is failing to meet the needs of birthing women in developing countries.

According to the international definition, a midwife is one who graduates from a program duly recognized in its jurisdiction. In the developing world, this generally means a two-year government training program. The women who attend these programs are usually young, often fresh out of high school, and have borne no children themselves. They are educated in an urban environment, then sent out to serve in a rural village, where they wear the white coat and expect respect from the townspeople for their professional, educated status. They usually work in the government-built clinic, but for an extra sum of money will sometimes attend a homebirth if they are called. These government clinics are usually underfunded and understaffed, and often the drugs with which they are supposed to be stocked are sold out the back door by clinic staff to compensate for the inadequate salaries they are paid. Recent anthropological ethnographies describe indigenous women in India, Mexico, Tanzania, Papua New Guinea, and elsewhere saying the same thing about the care they receive in these clinics: "They expose you, they shave you, they cut you, they leave you alone and don't come when you call, and they won't allow your relatives to be with you." Here is a highly representative quote from an article by anthropologist Pauline Kolenda describing birth in a hospital near a small village in India:

Before entering the hospital we have first to decide how much money we have to give. We are not admitted unless we first give them money. When the woman enters into the hospital, the doctor behaves rudely with her. Sometimes nurses beat her. They do not let close and affectionate relatives, who came from home with us, stand at our side. They themselves either do not stay near us. We wish that somebody hold us by the waist when pains come, but they do not do it. We have not even to moan, lest they talk sarcastically, make fun of us, which is very hurting, still we have to bear. If we moan too much, they may sometimes slap us. If we happen to say something, they retort by asking us whether they had invited us to come. "Why have you come then? You may go back home!" In hospital we have to lie down on the bed to get delivered. In the hospital they excise the vaginal wall with a blade for enlarging it. The body gets damaged unnecessarily. After delivery we feel terribly hungry, but we consider ourselves lucky if we get a cup of tea.

Not surprisingly, even though the governments of these countries have embarked on massive programs to bring birth into the clinics and hospitals, many rural women resist, choosing instead to birth at home with a community midwife. Officially labeled "traditional birth attendants" (TBAs) by WHO and UNICEF because they do not meet the international definition of a midwife, these community midwives are usually older women who have given birth several times and who have become midwives by being asked to attend the births of friends and relatives, slowly gaining first-hand experience of birth. Some of them undertake long apprenticeships with experienced community midwives, while others learn simply by attending births. From the point of view of villagers and townsfolk all over the developing world, the biggest difference between community and professional midwives is that community midwives are recognized by their community as midwives, while the professionals are often seen as young and inexperienced women who have to prove their worth to the villagers before they can be trusted.

When the professional midwives make a sincere effort to learn about and honor local customs and traditions, when they approach local people with an attitude of respect and demonstrate willingness to work with community midwives, honoring them as colleagues, this hierarchical system can function efficiently. In such cases, community midwives are generally very willing to advise women to go to the clinic or hospital in situations of risk, and will often accompany them there if they are sure of being well received and of getting adequate care for their clients. But when doctors and professional midwives approach the community with an attitude of arrogance, treat the "TBA" with disdain, and punish women who attempt homebirth by treating them badly when they do transport—an all-too-typical situation-a consensual agreement is made among mothers and their at-home attendants to avoid the clinic or hospital at all costs. Such a situation leaves the community midwife to cope with emergencies at home as best she can, often until it is too late to seek help. If disaster befalls, of course, it is she who takes the blame, when in fact the fault may lie in the dysfunctionality of the system.

Debora Barnes-Josiah and her colleagues have shown that in Haiti, as in other countries, a lack of confidence in available medical options was a crucial factor in delayed or never made decisions to seek medical care. They suggest that improving the quality and scope of care available at existing medical facilities will prove crucial in reducing needless maternal deaths. Likewise, Soheir Morsy has shown that in the rural area of Egypt she studied, traditional midwives were getting blamed for the high rates of maternal mortality, when in fact most of the deaths occurred among women who had given birth in the hospital as a result of complications from cesarean sections.

UNICEF, WHO and those engaged in implementing the Safe Motherhood Initiative have for two decades centered their efforts to reduce maternal and perinatal mortality in the Third World around "TBA training'—short, usually two-week-long courses taught by medical personnel, usually doctors, nurses, or professional midwives to community midwives. The purpose of these courses has generally been to educate TBAs about the risks that require transport and to improve their prenatal and maternity care. Of course, these courses too have been plagued by the same systems failures as the situations I describe above. Almost always, these courses are designed by biomedical personnel trained in biomedical institutions to think about and manage birth in biomedical ways. Very seldom do the "trainers" enter a community and spend time there learning about the indigenous birthways before they try to intervene. Rather, they attempt to educate traditional midwives in biomedical ways of thinking that are often totally inappropriate to local circumstances and realities.

For example, perhaps the local custom is to cauterize the cord with a candle flame after cutting it. Trying to replace that sustainable custom with Merthiolate in a place where supplies are scarce and Merthiolate unavailable or expensive is an unsustainable and inappropriate intervention, but one nevertheless that typifies this training approach. Far more seriously, TBA trainers often think their job is done if they succeed in educating midwives about the multiple conditions that are biomedically deemed to require transport to a clinic or hospital. The training complete, they leave, and the midwives get blamed if they do not transport for the risks they have just been educated about.

In Mexico and other countries, UNICEF has just discontinued funding for TBA training courses; since maternal mortality rates have not dropped after twenty years of TBA training, the conclusion is that they do not work. This conclusion is based on the assumption that mothers die because midwives give them inadequate care or fail to transport them in cases of need. But as we have just seen, sometimes it is the hospital that gives inadequate care. And often women in need are simply unable to reach the hospital. I remember well when Dona Nieves, a very short and very experienced traditional midwife from rural Oaxaca, Mexico, bravely stood up in the big auditorium in Mexico City in the middle of the Safe Motherhood Conference and said to all assembled:

Do not blame us for failing to transport women. We know when we should transport. But none of us own cars, nor do our clients, the buses run very irregularly, there is no ambulance service and if there were our clients couldn't pay for it, and the only taxi driver in our town charges far more than our women can pay. How then do you expect us to get our clients to the hospital in the city an hour away? No, we can't, we just have to do the best we can with no help from anyone. If you want me to transport women who need to go to the hospital, give me a car!

Another traditional midwife from Oaxaca, Dona Queta, described to me how a woman arrived to her tiny village high in the Oaxacan mountains many years ago. The woman, nine months pregnant, had tripped over a piece of wood and fallen hard. As she got up, she felt the baby move once, and then no more. Afraid that the baby had died, the mother walked for two days over the mountains to get to Dona Queta's home. By the time she arrived, said Queta, the odor around her was foul and it was obvious that the baby was dead. What to do? The woman had arrived alone and penniless. Dona Queta went to fetch the one doctor within 100 miles, but he was away on a trip. So she spent the next three days pouring liters of antibiotic herbs into the woman and praying for her life. Finally the woman went into labor and gave birth to the dead fetus, and then almost died herself from massive infection. But Queta persevered with the antibiotic herbs and other nutrients, the rituals, and the prayers, and two weeks later the woman walked home alive and well. When the doctor arrived back from his trip and found out what had happened, he scolded Queta for taking on the care of this woman, but had to back off when he realized that he had left her with no other choice besides leaving the woman to die on the road.

In an ideal world, the community midwife is the first line of care and is backed up by professional midwives, doctors, and the biomedical system. In the real world, often there is no backup or no way to get to it, and she must handle whatever comes as best she can. Clearly, the solution does not lie in giving her superficial training in biomedicine and expecting her to get women to the hospital when they need to go; rather, the solution must be found in a system-wide approach that requires as much flexibility of biomedicine as it does of the community midwife.

Consider for example the case of the Karimoja, a rural tribe of cattle herders in a remote comer of Uganda. British midwife Sally Graham went to Uganda some years ago to work in a public health clinic serving the Karimoja. Matemal mortality rates were high among this group, and Graham wanted to improve maternity care. She had been trained to think that the Western system was the best; nevertheless, it occurred to her that it might be a good idea to learn about the indigenous system before trying to change it. Slowly getting to know the local midwives, she eventually developed good enough rapport with them to suggest weekly meetings during which they could engage in mutual discussion of their different techniques. Graham discovered over time that these midwives had many skills; she came to realize just how many one day when a wizened older midwife came running out to stop Graham's Land Rover, waving her hands excitedly. She told Graham she was very sony that she would miss the meeting that day; she had to stay home as both her daughter and her daughter-in-law had given birth the night before to twins. One of the twins (the second to emerge) had been transverse, and she had to do an internal version to get it out. Both mothers, and all four babies, were doing well!

It turned out that the indigenous system didn't really need changing; the problem lay, as it so often does, in the interface between the biomedical and the indigenous systems. Karimoja midwives who transported were often rudely and dismissively treated, so they tended to hold on at home longer than was wise in order to avoid subjecting themselves and their clients to such disrespect. Graham's solution was to bring the clinic staff, two at a time, to the weekly meetings she held with the midwives. As the staff developed more respect for the community midwives, they invited them to the clinic for tours and get-acquainted sessions. Then when the midwives transported, their advice was respectfully solicited and listened to, and they were invited to remain with their clients and give labor support. This model of mutual accommodation, which Graham called the "partnership paradigm" was implemented over live years ago as a result of Graham's work and is still functioning well. (It reminds me of the models American direct-entry midwives often develop with doctors and hospitals in their communities. After years of mistreatment by hospital personnel, over time they earn their respect and develop good working relationships that exemplify Graham's partnership paradigm.)

All over the developing world, in spite of the massive dysfunctionality of most obstettic systems, it is possible to find models that work. The system that Dr. Galba Araujo developed in northeastern Brazil is one prime example. Araujo was the head of OB-GYN at a tertiary care center in Fortaleza, Ceara', Brazil. Concerned about high mortality rates in the rural regions his hospital served, he went out to the rural communities and asked the midwives what was needed. Their answer was that women needed a clean and safe place to give birth; most of their houses had dirt floors, and cleanliness was almost impossible to maintain. So he created a system of maternity care clinics in numerous rural villages. These centers were equipped with donations from local villagers: anything was welcome, from a fork to a plate to a chair. Each center was equipped with the hammocks in which local women preferred to give birth, and with the drugs and equipment that Dr. Araujo felt the midwives should have. He also created an efficient ambulance system, so that transport was readily available should the midwives call for it. Outcomes in these maternity centers were so good that Dr. Araujo began to study what the traditional midwives did, and ended up incorporating hammocks, more patience, and upright positions for birth into the hospital. This system functioned efficiently until his death; it has since been dismantled due to lack of support and interest on the part of the younger obstetricians who replaced him. While it existed, this system was an excellent example of Graham's "partnership paradigm" or what Brigitte Jordan has called "mutual accommodation" between biomedical and indigenous systems.

In 1978 with the publication of Birth in Four Cultures, Jordan issued a call for the replacement of top-down, culturally inappropriate, biomedically oriented systems with models of mutual accommodation like the ones Dr. Galba and Sally Graham created. But the worldwide hegemony of Western biomedicine has made this kind of mutual accommodation an elusive goal. Why bother to accommodate to a system you regard as inferior? Why not, as has so often been done, demand instead that the indigenous system change to accommodate biomedical ways of knowing and managing birth? The answer is quite clear of course: biomedicine is an inappropriate model for birth in any culture. It is too costly, too interventive, too drug- and technology-oriented, and does too much harm to mothers and babies for it to be a viable model to which developing countries should aspire. Nevertheless, because of the general global dominance of the West, the legacy of colonialism, and the dramatic successes of biomedicine, all developing countries do aspire to meet the standards set by Western medicine, standards that are inappropriate for birth and indeed for many of the world's health needs.

Consider the following description of a hospital birth in rural Papua New Guinea from the doctoral dissertation of Julia Byford, an Australian nurse-midwife who has become an anthropologist:

Mispa, a young woman of twenty, was admitted to the hospital this morning. She is seen by the Health Examination Officer, who does a vaginal exam and tells me that she is four to five centimetres dilated...and that she may commence a Syntocinon infusion...The labor room is small and the air conditioning unit mounted high on the wall belches cold air at us. When it is on the room gets very cold: when it is off the room gets hot and stuffy. There is a sink but no plumbing to allow it to be operational. There is no water at the hospital today anyway .... Mispa asks to sit on the floor and is given permission to do this, but as her labour progresses the nurse says she must stay on the bed so the staff can do their observations. She acquiesces and does not ask for anything else. Most of the time she is left alone. She has not eaten all day and only drunk a small amount of water. Her lips are dry and swollen. The staff do numerous vaginal examinations but none of them are recorded [so when a shift changes, another exam is performed]....

During the second stage of labour, every time Mispa has a contraction, the Health Examination Officer inserts a few fingers into Mispa's vagina between the perineum and the baby's head in order to stretch the perineum. Mispa finds this excruciating and tightens her grip on my arm...[After the birth] I am dismayed although not surprised to see that the baby is flat and pale and requires resuscitation. The HEO delivers the placenta by placing one hand on Mispa's abdomen and pulling on the umbilical cord with the other hand...As soon as the placenta is out, Mispa has a large postpartum hemorrhage. The HEO asks me to increase the intravenous infusion rate and then inserts her hand high up into Mispa's vagina and manually removes some retained placental pieces. This is done without explanation or anesthetic....

Perhaps the hardest thing for me to come to terms with is the lack of care offered to Mispa simply on a human level. She was never consulted, only told what to do and what not to do.... No one tended to her basic needs for food or fluids or inquired if she needed to go to the toilet. It was as if Mispa, the embodied person, did not exist. (Byford, 1999:186-190)

Unfortunately, this description is all too typical of hospital births in the developing world. Byford's dissertation describes the general poor health of rural Papuan women, who suffer from overwork, exhaustion, anemia, malnutrition, and a variety of diseases that result from their lack of access to clean water. She discusses research in public health which shows that the single most important intervention that colonial health services could have brought to PNG would have been an adequate clean water supply. But biomedicine, here as elsewhere, prevailed, and instead of investing in clean water, PNG invested in doctors and hospitals in the cities, and rural clinics like the one Julia describes above, which themselves do not even have running water most days of the week. This biomedical approach to health makes it appear that problems inhere in individuals and should be treated on an individual basis, patient by patient, hospital by hospital, obscuring the fact that the major causes of disease and death, among parturient women as among the general population, are structural and will benefit far more from large-scale systemic change than from diverting money into hospitals and clinics.

Given the difficulty of reaching the hospital and the poor treatment they receive once they get there, most rural Papuan women still choose to deliver at home—a trend paralleled around the developing world as rural women, disillusioned with clinics and hospitals, return to their community midwives and traditional birthways, leaving development planners to shake, their heads over this unaccountable unwillingness to use modern facilities, attributing it to ignorance and close-mindedness. On the contrary, birthing at home with a traditional midwife is often a considered decision that women make after weighing the risks and benefits of their options. For example, Morsy (1995:168) notes:

In villages of the Nile Delta, although modem medical facilities are available, women prefer to deliver at home with the assistance of a local midwife. This choice is not a result of reified "cultural attitudes" but a measured judgment about the inadequate health care extended to peasants and the urban poor in modem-health care settings.

Anthropologist Denise Roth has described this process of "measured judgment" in detail in her forthcoming book Bodily Risks, Spiritual Risks: Contrasting Discourses on Pregnancy in a Rural Tanzanian Community. Policy planners on high decided that each Tanzanian mother should have and be in charge of a card documenting the prenatal care she had received and her health and pregnancy condition. To make sure she carried the card, it was decided that she would have to show it in order to be admitted into the local clinic for labor and birth. In the city and small town Roth studied, this confronted each pregnant woman with a difficult decision: should she obtain prenatal care, which was costly, time-consuming, and often inadequate, in case she needed to go to the clinic during labor, or should she cast her lot with the traditional midwife, who was inexpensive and kind but would be unable to take her to the clinic without the card? Not surprisingly, many women after careful reflection chose the latter option; of course, if they then had problems during labor but could not go to the clinic, the traditional midwife would be blamed for the outcome.

Certainly, as Roger and Patricia Jeffrey pointed out in 1993, it is important not to romanticize indigenous midwifery and indigenous midwives; not all of them are skilled, not all of them give women good care; and some indigenous customs can be as harmful as many obstetric procedures. Those anthropologists most concerned about over-romanticization of the traditional midwife, including the Jeffreys, usually study birth in countries like India and Bangladesh, where women's status is very low. In a recent cross-national study, Shen and Williamson point out that low status for women directly correlates with higher maternal mortality rates. Where women's status is low, their nutrition is poor, their overall health is poor, and community midwives are less able to develop effective knowledge systems, as Rosario (1998) and others have shown. Where women's status is higher, community midwives are often able to develop long-standing and viable systems of indigenous knowledge about birth, as indicated in the growing body of ethnographic literature I cite in this article. This literature consistently indicates that community midwives (TBAs) usually give skilled and considerate care and remain, in many parts of the world, the only viable option for millions of women. As an anthropologist, I question the wisdom of dividing professional midwives and TBAs in this hierarchical, biomedically oriented way which allows government agencies and development planners to support one group while trying to exterminate the other. Can't a "real midwife" either be recognized by her government or by her community as such?

In sum, the present policy of separating professional from traditional or community midwives has led to midwives' integration into a hierarchical, intensely colonialist system that has doctors at the top, professional midwives in the middle, and community midwives at the bottom, with no power and very little government support. In this system, doctors have most of the power. Professional midwives, who are usually biomedically trained, often buy into this hierarchy, and work to impose biomedical models of birth on indigenous populations—a situation Australian midwife Leslie Barclay calls "midwifery hegemony." The surest sign that such a system is in operation is when women who have birthed upright for countless generations are suddenly told by the midwife or doctor to lie down. Countless pages of scientific evidence now document the efficacy and superiority of upright positions for birth! Yet biomedicine in its arrogance insists that its way is best, and around the world is still working to eliminate the few remaining viable indigenous systems of birth, teaching even community midwives to make women lie down for birth, and replacing home with hospital wherever possible.

In rural Thailand, for example, anthropologist Andrea Whittaker (1999) is documenting the escalating biomedicalization of birth. As more and more rural women spend large sums of money to have the more prestigious hospital births, the community midwife's role is being reduced to postpartum care. Ironically, at the same time as indigenous birthways are rapidly vanishing in the rural areas, in Thai cities, where birth has long been biomedicalized, an incipient alternative birth movement finds some professional midwives beginning to attend the births of middle-class women at home (Andrea Robertson, personal communication).

In the words of songwriter Joni Mitchell, "Don't it always seem to go, that you don't know what you've got till it's gone?" Must we lose the viable indigenous birthways that still exist before we rediscover how valuable many of them were? Must the professionalization of midwifery mean its colonization by biomedicine? Or can professional midwives reclaim their autonomy, foster the globalization of the midwifery model of care in culturally sensitive ways, work in tandem and cooperation with traditional midwives, and become worldwide the agents of mutual accommodation and positive change?

Robbie Davis-Floyd, PhD, is a Research Fellow in the Department of Anthropology at the University of Texas, Austin. She is author of Birth as an American Rite of Passage (1992), coauthor of From Doctor to Healer: The Transformative Journey (1998) and has written and edited numerous other books and articles. She lectures nationally and internationally, and is currently at work on Mainstreaming Midwives: The Politics of Change.

References Cited and Suggested Reading

• Asowa-Omorodion, Francisca I. (1997). Women's perceptions of the complications of pregnancy and childbirth in two Esan communities, Edo State, Nigeria. Social Science and Medicine 44: 1817-1824.
• Barclay, Leslie M. (1998). Midwifery in Australia and surrounding regions: Dilemmas, debates, and developments. Reproductive Health Matters 6(11): 149-155.
• Barnes-Josiah, Debora; Myntti, C. & Augustin, A. (1998). The three delays as a framework for examining maternal mortality in Haiti. Social Science and Medicine 46: 981-993.
• Byford, Julia. (1999). Dealing with Death Beginning with Birth: Women's Health and Childbirth on Misima Island, Papua New Guinea. PhD dissertation, Dept. of Anthropology, Australian National University.
• Davis-Floyd, Robbie & Sargent, Carolyn. (1997). Childbirth and Authoritative Knowledge: CrossCultural Perspectives. Berkeley: University of California Press.
• Graham, Sally. (1999). Traditional Birth Attendants in Karamoja, Uganda. PhD dissertation, South Bank University.
• Jeffery, Roger & Jeffery, Patricia M. (1993). Traditional birth attendants in rural North India: The social organization of childbearing. In S. Lindenbaum & M. Lock (Eds), Knowledge, Power, and Practice: The Anthropology of Medicine in Everyday Life, pp. 7-31. Berkeley and London: University of California Press.
• Jordan, Brigitte. (1993 [1978]) Birth in Four Cultures: A Cross-Cultural Investigation of Childbirth in Yucatan, Holland, Sweden and the United States. (4th ed.) Prospect Heights, Ill: Waveland Press.
• Kolenda, Pauline. (1998). Fewer deaths, fewer births. Manushi 105 : 5-13.
• Morsy, Soheir. (1995). Deadly reproduction among Egyptian women: Maternal mortality and the medicalization of population control. In F. Ginsburg and R. Rapp (Eds), Conceiving the New World Order: The Global Politics of Reproduction, pp. 162-176. Berkeley: University of California Press.
• Rosario, Santi. (1998). The dai and the doctor: Discourses on women's reproductive health in rural Bangladesh. In Maternities and Modernities: Colonial and Post-Colonial Experiences in Asia and the Pacific, pp. 144-176. Cambridge UK: Cambridge University Press.
• Roth, Denise. (2000). Bodily Risks, Spiritual Risks: Contrasting Discourses on Pregnancy in a Rural Tanzanian Community. Ann Arbor: University of Michigan Press. In press.
• Shen, Ce & Williamson, John B. (1999). Maternal mortality, women's status, and economic dependency in less developed countries: A crossnational analysis. Social Science and Medicine 49: 197-214.
• Whittaker, Andrea. (1999) "Birth and the Postpartum in Northeast Thailand: Contesting Modernity and Tradition," Medical Anthropology 18(3): 215-242.

If you enjoyed this article, you'll enjoy Midwifery Today magazine! Subscribe now!