The shifting axes of marginalities: the politics of identities shaping women’s experiences during childbirth in Northeast India

Abstract Institutional births in India, including the north eastern state of Assam, have increased steeply in the last decade such that 71% of all births now occur in facilities. Most analyses of disrespect and abuse during childbirth have largely framed the problem within a binary that juxtaposes all users of services in one category, subordinate to institutions and institutional actors. This commentary explores whether a different analysis is possible within a relational context where citizenship itself is graded, and not all marginal groups experience either the same form or the same intensity of mistreatment. Employing a historical lens including examining relations between non-elite groups, current discriminatory state policies and practices, and deepening conflicts over scarce resources, this commentary presents a more localised and granular understanding of how disrespect and abuse may manifest in institutional births in Assam. Experiences of disrespect and abuse during childbirth are mediated by axes of marginalities that are dynamic and non-isomorphic, shaped by state policies, the everyday practices of the citizens, the differential and unequal relations between the state and multiple marginal groups of citizens, and between citizens themselves. Reframing marginality in this way may lend itself to identifying sources of inequities that emanate from both within and outside of health systems, allowing for more sophisticated explorations of disrespect and abuse. This may help improve health systems to ensure that experience of childbirth is more humane, safe and respectful, independent of women’s social identities and their locations in the larger political economy.


Introduction
Women's experiences of disrespect and abuse during institutional childbirth encompass actions such as physical, verbal and sexual abuse, discriminatory behaviour, not being offered a choice of birthing positions or pain relief and other systemic and interpersonal practices. 1 However, the risks of encountering disrespect and abuse during childbirth are not shared equally by all women. Some studies indicate that women who are marginalised due to their ethnicity, religion, caste, race, class, gender, and sexual orientation are disproportionately affected by such mistreatment and poor quality of care. 2 In this commentary, I use the illustrative example of Assam, a state in the northeast of India with a heterogeneous population, a complex colonial history and post-colonial politics, and the highest Maternal Mortality Ratio in India. The question I pose is whether the experience of disrespect, abuse and discrimination during childbirth is the same or different across marginalised groups such as Bengali Muslims, Adivasis (tribal communities, originally from eastern and central India), and indigenous tribal populations. This issue is significant since, in the last decade, there has been a big push towards institutional births in India (and also in Assam) to address maternal mortality.
Institutional births have increased in the state from 22.4% in [2005][2006] to 70.6% in 2015, of which over 60% are in state facilities .
Historically, the relationships between these communities have been tenuous, not only due to ethnic, linguistic, religious and cultural differences, but also due to struggles over scarce resources, most notably arable land in a region affected by denudation from seasonal flooding. In recent times, these relationships have worsened due to state policies that give preferential treatment to some groups, while actively discriminating against others. For instance, the formulation of the National Register of Citizens in 2018, the only one of its kind in India, tries to establish the right to residence based on documented proof of citizenship in the state of Assam. Thus, not all of these marginal groups are similarly or equally disempowered in Assam. A conceptual framework anchored in an understanding of graded inequalities of power that change over time may help unpack the different sources of inequities that influence women's experiences during childbirth, with implications for the measurement, documentation and analysis of disrespect and abuse. This has the potential for improving institutional care, not just in Assam or in India, but in other complex regions with differentially marginalised populations.
The issues raised in the commentary have been addressed by two related areas of scholarshipintersectionality in health and the political economy of development. The former examines the ways in which markers of identity interact with interlocking systems of privilege and oppression at the macro level, affecting health outcomes. 3 For example, a study examining the combined and independent impacts of gender and class in rural Southern India found that non-poor women and poor men achieved similar treatment outcomes for chronic long-term illnesses through different processes: the former leveraged their economic advantage to overcome their gender disadvantage, while the latter did the opposite. 4 Within the political economy of development, earlier scholarship had merged all marginalised populations (e.g. peasants, working class) within a single category of the subaltern. 5 Later, scholars concerned with disadvantages stemming not only from class but also caste, tribe, and gender, have argued that the category of the marginal is not singular. 6 This commentary uses the concept of what I term "the shifting axes of marginality" to argue that over a period of time, the positions of erstwhile non-dominant groups may change based on negotiations with the state and changing relationships with other non-elite groups. For groups that may become relatively more empowered, there are likely to be positive outcomes across domains, such as livelihood, education and health. For groups that become relatively more disempowered, this will be less so. These shifting axes may have implications for how vulnerable and marginal groups of women experience disrespect and abuse during institutional childbirth. This, in turn, raises concerns about how to understand care and abuse in nuanced ways and develop sensitive interventions to ensure respectful maternity care, independent of women's positions in the larger political economy.

Context of disrespect and abuse
Demographic governance mediated through technology, surveillance, enumerative exercises, forced birth control and pronatal policies of the state, are not new phenomena. 7 In the context of disrespect and abuse during childbirth, state power continues to be deployed through technology, institutions and entities such as doctors, nurses, bureaucrats and administrators, and professional midwives, who are sometimes of a different (higher) social class than the patients they serve. Coovadia et al find that a history of colonisation, apartheid and a discourse rooted in patient inferiority have contributed significantly to the perpetuation of disrespect and abuse of women within health systems. 8 One of the earliest analyses of this problem in South Africa finds that abuse of pregnant women by nurses took two forms: as a response to an event, such as perceived non-compliance, and as ritualised abuse, especially in the absence of a competing discourse of care. 9 The authors identify a complex set of factors, including systems of accountability (or their absence) within institutions, the deliberate creation of social distance by nurses consolidating their middle-class identities, and the need to assert control, which may inhere in the technocratic model of childbirth. 10 In the Indian context too, disrespect and abuse during childbirth are not uncommon, especially in state facilities, and have been reported by journalists as well as academic scholars. Empirical work in Assam finds several instances of abuse in state facilities in rural Assam and argues that such violence is gendered " … and intersects with other axes of structural inequalities, such as caste/ethnicity, class and region in India, producing a negative and often violent experience of pregnancy and childbirth in India". 11 Using a mix of clinical observations and quantitative methods, a study based in state and private facilities in urban and rural Uttar Pradesh, North India, reveals that while all women experienced at least one type of mistreatment during childbirth across facilities, women who were either Scheduled Tribes or Scheduled Castes, as well as multiparous women, reported greater and more intense forms of mistreatment. 12 The latter include actions with significant possibilities of harm, such as extreme fundal pressure, using dirty rags to wipe the perineal area, physical abuse, and sexually humiliating comments. Variations in mistreatment across different social groups are also confirmed in other studies in India. 13 However, disrespect and abuse in childbirth are likely to be underreported by marginalised women because of low expectations of care in state facilities and the propensity to self-blame for the poor treatment they receive. 14,15 Thus, gender along with other axes of inequities disempowers these women because they often possess the least social, economic and cultural capital to exercise agency during such medical encounters.

Hierarchies of citizenship
Although a leading global tea-producer, Assam is among the poorer Indian states. It has a population of 31.2 million, with great ethnic and religious diversity, comprises Assamese Hindus (the majority), along with significantly large minority groups of Bengali Muslims and Bengali Hindus, indigenous tribals, Adivasi tribals and a small minority of Assamese Muslims. 16 Assam has a large internally displaced population on account of annual flooding and decades of ethnic conflict, the latter an outcome of contemporary and colonial policies of the state.
British colonial policies, including the marginalisation of indigenous tribal populations, and the policy of importation of Hindu Bengali middleclass professionals and Muslim Bengali peasants from East Bengal (present day Bangladesh), alongside Adivasis as indentured labourers from East and Central India, fomented both anti-outsider and anti-Bengali feelings in the state. 17 After India's independence, other policies have reconfigured state-citizen relationships and created a hierarchy in citizenship that has exacerbated difficult relations between several marginal groups. Exclusionary rules guarantee preferential political, economic, social, and forest rights to some groups.* , 18 Marginalisation of Bodo tribals, along with competition for increasingly scarce arable land between Adivasis and Bengali Muslims, led to decades of violent separatist struggle. 17 Simultaneously, the state has introduced requirements that make proving citizenship difficult: for example, through the National Register of Citizens for Bengalis, making statelessness a significant threat for Bengali Muslims in the region.
For historical reasons and a paucity of land in Assam, many Bengali Muslims have settled in the almond-shaped islands of the lower Brahmaputra river (char region) areas which are subject to severe annual flooding, exacerbated in recent years by earthquakes and climate change, and with a near absent public health infrastructure with boat clinics offering basic health services run by nonprofit organisations. When Bengali Muslims move to less flood-prone areas of Assam, to diversify their livelihoods or due to homelessness from floods, they are labelled as "illegal Bangladeshis" or "foreigners", and subject to multiple forms of discrimination and violence. 19 Between 1970 and 2015, and notably even after the formation of tribal autonomous areas, conflicts between Bodos and Bengali Muslims as well as between Bodos and Adivasis led to over 4,000 deaths and nearly a million people displaced across all three groups. The majority of victims have been Bengali Muslims, followed by Adivasis. The worst of these was the Nellie massacre in 1983 where nearly 3,000 Bengali Muslims were hacked to death in a single night in response to the central government granting voting rights to them in the state elections. However, the perpetrators were never arrested. 20 On the other hand, Adivasis, most of whom live and work in the tea plantations in Assam, have continued to agitate for "indigenous" or scheduled tribe (ST) status in the state, which they are entitled *These rights were granted to Karbi and Bodo tribals (considered "indigenous" by the state), who constitute 12-15% of the state's population, but not to Adivasi tribals, who comprise 25% of the state's population, because they are considered "outsiders". Bodo demands for autonomy originated from legitimate experiences of their linguistic and cultural marginalisation and neglect by colonial and successive state governments and culminated in the creation of autonomous Bodo districts through the Assam Accord of 1985.
to outside of Assam and which offers them more social protections. It is worth noting that MMR and other health indicators are worst in the Adivasi areas and in districts where Muslims constitute the majority of the population. 21  The multiple axes of disempowerment The precarity of being a Bengali Muslim in Assam The districts with majority Bengali Muslims have some of the worst health indicators, 21,22 partly due to a systematic underinvestment in statefunded institutions and partly due to structural inequities experienced by the population. Unpublished data from our study 11 suggest that ASHAs (Accredited Social Health Activists), encounter multiple challenges. First, they have to negotiate intrapartum care for pregnant women in health facilities, where they are often mistreated, possibly because of being both poor and Muslim. Second, ASHAs struggle to prevent early marriages that increase the risks of obstetric complications and maternal deaths, while simultaneously reinforcing negative stereotypes against the community as "backward with unfettered fertility". Third, they encounter more cases of non-payment of statemandated cash benefits to Muslim mothers compared to other groups under the Janani Suraksha Yojana (JSY), which is a pan-Indian conditional cash transfer scheme that provides financial incentives to pregnant women and health workers for institutional births in state facilities. Finally, several ASHAs (in these districts but also elsewhere) are often not paid their wages for several months. This exacerbates the existing economic precarity that the ASHAs live with and also diminishes their ability to negotiate (as women) with their families their right to be employed.
Discrimination against Muslim women in public healthcare settings has also been reported in other parts of India. A study in Mumbai found that culturally derogatory terms were used against Muslim women. They were perceived as being troublemakers, staff refused to learn their names, and substituted Hindu names, and many women reported being humiliated for having more than two children and felt they were spoken to differently from non-Muslim patients. 23 However, given the colonial history and contemporary politics, Bengali Muslims in Assam are especially vulnerable. Some Bengali Muslims used the word "sub-human" to describe their experience of childbirth and accessing health care. 11 This perception is reinforced, when we examine "native Assamese" and mainstream discourses that often label Bengali Muslims as "illegal immigrants". Bengali Muslims are blamed for "large" populations, unregulated fertility, and for poor health statistics in the areas where they live, and are perceived as being cunning and more adept at accessing state entitlements than the "native" population. This view is unfortunately also held by some health providers, many of whom belong to the dominant or elite groups in Assam. The use of the word "immigrant" and concerns with "excess" Muslim fertility are instructive, though the latter is not new; the politicisation of fertility to achieve non-demographic agendas has been debated in India for decades. These issues raise several questions when researchers find more incidents of mistreatment of such groups. If many Bengali Muslims are poorly treated in state facilities, owing to prejudices held against them, how likely are they to use state facilities? Even if they were to do so, what would be the quality of the care they receive?
Othering Adivasis through home births When Adivasi women have home births, this is seen as a repudiation of state-sponsored development agendas, such as promoting hospital births, and reinforces negative stereotypes of Adivasis as "backward" and resistant to development initiatives. However, there are multiple legitimate reasons for this choice. † For historical reasons, including lack of educational opportunities in the tea garden areas, the majority of Adivasi men and women have continued with colonial occupations such as work in tea plantations. They have few social protections and are routinely exploited by tea garden owners, and receive only half the state-mandated minimum wage because of archaic colonial rules that still govern labour laws in the tea plantations of Assam. 24 A visit to a † More than 50% of Adivasis employed in tea plantations are contractual employees with few labour rights and are addressed by a derogatory term, Faltu meaning useless in Assamese, Bengali and Sadri, the language spoken by Adivasis in Assam. government health facility, often several kilometres away, entails a loss of wages. On arrival at the facility, significant out of pocket expenses are incurred: cost of tests (50-70 Rupees), bribes of hospital staff (20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30), tea and transportation for the ASHA and for themselves, adding up to 150 Rupees (2.5 USD), unaffordable for a family with average daily earnings of 121 Rupees (1.9 USD). ‡,25 Many pregnant women do heavy labour right until they deliver and return to work within a few weeks of giving birth. Malnutrition and anaemia are extremely high, linked to high levels of poverty, and the colonial practice of using substantial quantities of salt to make the poorest quality of tea palatable often increases the risks of hypertension and preeclampsia for Adivasi women. § Further, in many Adivasi communities, childbirth is often viewed as a cultural rite of passage where the presence of families is preferred over health workers, who tend to be non-Adivasis. At the tea estate hospital, the conditions can be worse than delivering at home. Kalita's study set in a large tea estate of Sonitpur District reported a lack of basic equipment and/or staff, with no supply of even the surgical thread required for suturing tears during childbirth. Even when Adivasi women give birth at the hospital, they are severely reprimanded by staff for not bringing clean or new clothes for the baby, often made impossible due to their extreme poverty and long working hours. The combination of structural factors, poor experiences at the hospital, including being humiliated, and the absence of amenities discourage institutional births in these areas. These experiences of Adivasi women, however, may not be unique to India. It may parallel the treatment of other indigenous groups in the Americas and in Australia.
The popular perception of Adivasi women, including the view held by Bodo or Karbi tribal women, constructs them as backward, with weak bodies, drinking alcohol even when pregnant and with high maternal death rates. These perceptions follow Adivasis even if they live outside of tea gardens. Home birth is viewed as strictly Adivasi practice by these other tribal groups in that they feel that they have become "one of them" if birth happens at home. 11 Thus, Adivasis like Muslims face a dual burden of marginalisation; marginalisation by the state which denies them their rights in Assam, and by other citizens, including other tribal groups, who may be non-dominant, but are not equally marginal and are differently discriminated from them.

Tribals as ignorant
Indigenous tribes in Assam and elsewhere in the northeast are often imagined as recalcitrant, disobedient and savage-like. 17 Their experiences with institutions sometimes reflect these perceptions, which together with their often low socioeconomic status, subjects them to disrespect. While many of these women actively use state facilities and private care if they can afford it, indigenous women and their families are often scolded by doctors and nurses for screaming too loudly during labour or asking too many questions about the care they receive, a common form of abuse during childbirth that is reflected in the experiences of women with low levels of education or in poverty. [11][12][13][14][15] It is possible that the discrimination that indigenous tribal women encounter, emanates more from the state, than from their fellow citizens, unlike their Adivasi or Bengali Muslim counterparts, who bear the burden of not only state-sponsored mistreatment, but also prejudices held by their fellow citizens.

Conclusion
Most analyses of disrespect and abuse during childbirth, particularly in state facilities, have largely framed the problem within a binary that juxtaposes all users of services in one category, who are subordinate to institutions and institutional actors. This commentary explores whether a different analysis is possible within a relational context where citizenship itself is graded, and not all marginal groups experience either the same form or the same intensity of mistreatment. Here, a deeper examination of historical relations between non-elite groups, current discriminatory state policies and practices, and deepening conflicts over scarce resources allowed for a more localised and granular understanding of how disrespect and abuse may manifest ‡ Deep J. Kalita, a postgraduate student in Public Health at the Azim Premji University, conducted an eight-week ethnographic study, under the supervision of the author, supplemented by surveys with 60 women among Adivasis who had given birth in the last two years in a large tea estate in Sonitpur District with a population of 6700 Adivasis. It is worth noting that a few years earlier this was one of the areas affected by Bodo-Adivasi conflict and subsequently many Adivasis were displaced. § Reported to Kalita by the doctors serving this population.
in the context of childbirth in Assam. We observe that experiences of disrespect and abuse are shaped not only by state or institutional actions, but also the perceptions of marginal groups including prejudices that certain non-dominant groups hold of other non-dominant groups. This approach opens up several avenues for future research in the area of disrespect and abuse. First, can quantitative instruments be made more responsive to this reality, by capturing the multifaceted and dynamic nature of discrimination, instead of homogenising all marginalised groups into a single category? Second, sub-standard and disrespectful care may be considered adequate because users feel they do not deserve better treatment due to their social position, a sentiment that is reinforced through popular and dominant discourses and views of other marginal groups. Thus, enumerating disrespect and abuse in childbirth through self-reported measures alone complicates the picture, since marginal women are more likely to experience disrespect and abuse and yet they may be less likely to report it. Third, past historical injustices may lead to ascription of intentionality to experiences of disrespect and abuse, even if the harm was not deliberate. This, of course, does not undermine the harm done to women when they are extremely vulnerable, but does raise the question of whether future instruments and narratives of poor care during childbirth can untangle these complex, but analytically and practically important differences. Perhaps we could consider the axes around which marginalities pivot as dynamic and non-isomorphic, shaped by state policies, the everyday practices of the citizens, the differential and unequal relations between the state and multiple marginal groups, and between citizens themselves. This conception of marginalities may lend itself to identifying sources of inequities that emanate from both within and outside of health systems, allowing for more sophisticated explorations of disrespect and abuse. This is necessary to improve health systems so that the experience of childbirth for women is more humane, safe and respectful, independent of their social identities and locations.