AE | Volumes | Asian Ethnology 76 (2)

A Narrative of Saṭvāī Affliction in Rural Maharashtra

Deepra Dandekar



This article contextualizes a narrative of ritual affliction from rural Maharashtra within women’s experiences of gendered, clan, and caste oppression. A case study of affliction demonstrates and exemplifies how women are socially and ritually categorized as failures in their marital duty (dharma) of being good mothers and wives across caste, through the worship of a popular village goddess, known as the saṭvāī, an ambiguous figure who presides over childbirth in large parts of Maharashtra. Saṭvāī worship produces two social categories of women: 1) mothers of sons, an identity that places women in an esteemed position within patriarchal hierarchy, and 2) the subaltern position of an impure and “child-eating witch,” a social condemnation and ritual diagnosis that childless women come to share. The article seeks to question the postcolonial discourse, which describes native religious rituals as healing and foregrounds women’s micro-narratives of affliction that subvert ritual descriptions as restorative, revealing these to be located within dominant caste interests and clan consensus.

The healing potential of South Asian rituals is often theoretically described, explained, and bolstered with thick ethnographic descriptions. What is usually not described or is viewed only in terms of what it represents for healing rituals, is affliction itself. This interested me because I was engaged in exploring community health and how healing at rural shrines was contingent on a diagnosis that located the cause of illness in forces outside of human control: in the victim’s village, or her entire body politic, with all its interlinked social and political networks. Deities were said to assume control over evil afflictions and illnesses as a force other than the victim. They provided social recognition to the sufferer and thus accorded her with victim status that served to vindicate her.

However, as I undertook fieldwork with women in reproductive-health distress in various villages of Pune district (especially around Ambegaon), my notions about healing were to undergo drastic change. In my field area, being diagnosed as “afflicted” (socially excommunicated) for child-loss after suffering reproductive crisis, or being declared non-afflicted or “healed” (socially reintegrated into marital clans) after conceiving pregnancies depended on women’s social and personal status. Therefore, inhabiting the two social categories for women of “mothers” or “witches” depended on how they personally fared within their marital clans and according to patriarchal and agrarian caste rules in the village that were upheld by goddess rituals and diagnostic paradigms. This led to questions of whether they could or should be accorded victim status (blamelessness) for childlessness at all, as many who fared badly in their inter-relationships with marital clans received ritual affliction diagnoses at the very first instance of reproductive health crisis, and could never be vindicated or healed even after they became mothers. Women therefore often played a secondary role in relation to their marital clans, which played the decisive role in according them victim status based on intra-clan dynamics, the birth of sons, and dominant caste-consensus in the village. The village in turn interpreted the village goddess’s verdicts about affliction and healing, with the help of goddess mediums, who inhabited lower feudatory positions in the village. Before healing actually commenced, therefore, the social and personal status of those women ritually diagnosed as afflicted had to be first identified as important enough to be accorded victimhood.

The village goddess, who presides over reproductive crisis and both childbirth and child-death in most parts of rural Maharashtra, is known as the saṭvāī. I will provide a brief overview of the saṭvāī goddess below. This will be followed by a narrative of saṭvāī affliction that is posited here as an example of how women’s accounts of affliction do not necessarily constitute an understanding of ritual healing, but rather question it. This narrative will be followed by an ethnographic account of how saṭvāī affliction is diagnosed and a concluding segment that analyzes afflictions as being discursively different from ritual healing.

The Saṭvāī

I was familiar with the popular childbirth goddess, the saṭvāī, even before my research with afflicted women in rural Maharashtra began. References to the saṭvāī are widespread in the Marathi language, wherein she is mentioned as inauspicious, being sometimes referred to as unmarried or without paramour, and the writer of children’s fates on the fifth night/sixth day after childbirth. She is also foregrounded in the grey literature as a malevolent presence, accompanying the auspicious goddess Jīvantikā (cf. Śarmā 2012, 475–78). The prevalence of the saṭvāī as a rural goddess presiding over reproduction in Maharashtra has been pointed out by many other scholars, such as Feldhaus 1995a, Kosambi, 1962, and Dhere and Feldhaus 2011, even as the goddess has been associated with many varieties of childbirth and fertility goddesses, such as the maulyā, the bhivayyā, and the sātī āsrā in Maharashtra. These figures preside over women’s fertility, childbirth, and water-bodies that are conceptually interlinked with tantric notions of female reproductive fluids (Feldhaus 1995a; Feldhaus 1995b; Gordon-White 1996). At a popular level, the saṭvāī has further been associated with the mātrikā, Ṣaṣṭī, the saptamātrikās, yōginī worship, and disease goddesses such Ṣītālā, who is a very popular village goddess in Maharashtra. She is also associated with other deities, such as Mari-āī, Renuka-āī, and Yellammā.

Besides all these associations, the view of the saṭvāī in the area where I researched her cult demarcated her primary nature into that of an ancestral spirit. In contrast to her reputation as a goddess that rebelled against husbands, paramours, and marriage (and was therefore somehow a-marital and a-reproductive), women in my field area described her as a collective of married women’s ancestral souls, who had died while in labor to save their newborn sons, thus upholding their childbirth duty to their marital clans1. When I asked my village respondents about this contrast, they complained of how the saṭvāī’s reputation of being an unmarried and wild goddess was a Brahmin allegation, meant to denigrate non-Brahmin rural goddesses and women worshipping them as women whose motherhoods were aggressive and whose marriages were unchaste. My respondents, on the other hand, reiterated that the saṭvāī goddess was married and valorous, just like her followers, who despised menstruation only because the goddess felt that women should be pure and abhor witches and witchcraft in the village, since witches were considered to manipulate menstrual blood, known to contain malevolent powers. But the goddess was also considered to be malevolent and angry, and it was believed that she made women give up their possessive love for children, constructing this kind of motherhood (possessive love) as impure, because she herself could not enjoy motherhood and had died in childbirth. The saṭvāī, therefore, represented a type of reproductive function as a goddess whose shrines were located at physical village boundaries.

Each clan in the village near Ambegaon where I collected most of my data had its own saṭvāī and performed its childbirth rituals on the fifth night and twelfth day in her name, facing the direction of the main regional shrine2 that organizes the village and the region into many spatial hierarchies Dandekar and Dandekar 2011. Clan members also undertake a pilgrimage to her shrine in order to dedicate the first hair of their newborns to the goddess or sometimes dedicate this hair to the village saṭvāī in the name of their own clan saṭvāī, whose shrine may be located far away. Although the saṭvāī represents a spatial relationship simultaneously suggesting female motherhood duty and childbirth, it is a relationship established by rural dominant-caste agrarian clans with an influential rural deity whose clan tutelage they accept. There are, however, certain specific features that distinguish a village goddess as the saṭvāī as well. While she is known to consist of multiple numbers (usually an entourage of a pairs of seven sisters), as well as many forms typifying various castes inhabiting and representing various eco-zones, her most important criterion is her role as a boundary sentinel separating the dominant agrarian castes from the village’s lower castes. The saṭvāī shrine therefore marks the dominant-caste boundary of the village and produces caste and gender relationships across it by embodying its purity divide against the lower-caste inhabitation at the village periphery, and afflicting dominant-caste women, who physically cross this purity line with reproductive crisis and childbirth failure. Dominant-caste women who cross the purity divide suffer miscarriages as a result of their indiscretion and “impure” action of having frequented the goddess’s shrine at the village’s physical boundary during menstruation, and are thereby punished for adulterating their caste purity by exposing themselves to lower-caste spaces and thereby endangering the chastity of their marital clans. Their actions are considered to be tantamount to witchcraft, which is understood as women’s crimes against their marital upper-caste clans and castes. They are accused of consorting outside the village periphery with lowly castes that are associated with cremation grounds, garbage heaps, and dead bodies, which also exposes them to contaminating bodily fluids, such as menstrual blood near goddess shrines.

Shruti and Aparna

I originally began my research in villages around Ambegaon, Pune district, where I organized group discussions about the goddess saṭvāī with women. These snowballed into a series of group discussions in various surrounding villages, including the village where I was staying (Ghodegaon). I soon realized that neither the women within these groups, nor the goddess I was researching, were static. Neither was the concept of region. Rather, the women, the goddesses, and the region of my research had extremely fluid boundaries that were mobile and dynamic. Marriage in Maharashtra (like most other places in India, as shown by Palriwala and Uberoi 2008) is virilocal. Marriage was thus one of the greatest reasons for women’s migration into and within the region of my research.

Women within my initial discussion groups were mothers who brought their children along. These women were more interested in performing fertility rituals and projected themselves as normal, as they discussed domestic duties and child-rearing matters while playing with their babies. Asking me if I had had a child, they instructed me to have one child as soon as possible, instead of writing my dissertation on goddesses and childbirth. They warned me about the perils of childlessness, then provided me with interesting information about taboos concerning the saṭvāī and her rituals. But this group gradually grew disinterested in me as time passed, and they did not have much more to contribute thereafter. They only intermittently asked me if I had conceived a baby, when I would ask them whether I could attend saṭvāī rituals in their homes when someone in their clans gave birth (most of them declined outright or accepted, but then misinformed me about the time). This confused me initially, but I realized later that this was because I was childless and considered inauspicious myself.

The exclusion described above turned out to be a blessing in disguise, since it provided me with the opportunity for a more intense set of group discussions in and around the villages I was researching. These were groups of women, who after understanding that I was childless myself, included me by default in their groups. When they met, they recounted their affliction diagnoses, personal narratives, and stories of child-loss, assuming that I would participate and recount my experiences of the same. While being included in these ritual groups was interesting for me in the beginning, women soon asked me to perform saṭvāī rituals as well (Dandekar 2014; Dandekar 2016). I primarily performed rituals to equalize power with my respondents and not seem too distant as a Brahmin researcher from an urban area. These group exchanges catapulted me not only into a world of their personal stories of domestic violence suffered at the hands of marital clans for having suffered child-loss, but also into their dreams that they believed were being sent to them by the goddess through mechanisms of ritual affliction. These dreams became infectious within various sharing circles and marked the afflicted status of women who were under the goddess’s ritual control, though not every woman had these dreams (the women in the case studies I am sharing here did not, for example), which were often very traumatic. This sharing in smaller groups often introduced newer members such as relatives or visitors (like myself) to the group. Some went visiting to other villages with relatives and friends, who were women that had married and migrated to the region from other areas. Sharing dreams and affliction stories and accompanying each other on ritual sojourns became ways for women to recognize each other’s stages within the prototypical affliction narrative.

The groups and discussions about affliction narratives and the goddess, therefore, existed before I entered the scene, and every childless woman like me entered their circle just as I had. I had simply been included within their group naturally, as one among other childless and afflicted women, which soon disabused me of my sense of power about forming any discussion group of my own about my research. My writing about it, instead of having a child or being able to have a child, was simply viewed as a further expression of my affliction, and because I performed saṭvāī rituals too, my inclusion in the group was confirmed. During the time I worked on the goddess, I did not just conduct fieldwork; my relationship with women inhabited a certain gendered domain characterized by affliction, as I was considered by countless women to share their status as they casually discussed affliction symptoms and dreams with me, assuming that I belonged, even though I held a pen and paper in hand. However, I also received a considerable amount of friendship from afflicted women, but only as an afflicted woman myself (though I was never subjected to a ritualist). This allowed me many personal invitations to women’s homes to talk about their stories. I also tried to dissuade the women against the goddess’s ritual diagnosis narrative, but I had to be careful about not antagonizing their clans. Many personal stories were, however, fragmented or ended badly, and I spent less time in my field area toward the end of my research to decrease my personal fear about the goddess and her affliction dreams. I did, however, honor many personal invitations extended by my respondents to visit their natal villages in Marathwada, to study shrines and goddesses and record narratives at a later date. But these private discussions that took place with friends later were more concerned with gossip about other village women and allegations about their falsehood in various ritual sessions, both public and private. Personal narratives were only further shared when women fell very ill or when clans required some help out of me, such as in the case of this article’s narrative. In terms of ritual relationships, it took me a long time to extricate myself from the shared world within which afflicted women’s friendship had presented my own childlessness.

Sharing and group discussions mostly took place under the pretext of singing devotional songs, as some women even knew some traditional dance steps that were typical of their rural natal regions. None of them would have ever accepted these group discussions as being defined as spaces for sharing stories about afflictions, dreams, domestic violence, or saṭvāī rituals. Nor would they accept them as giving advice to others whose symptoms were discussed. A fair number of women belonged to different parts of Maharashtra, such as Marathwada (including Latur, Parbhani), Northern Maharashtra (Khandesh, mostly Bhusawal, Jalgaon, and Nandurbar), and the more western parts of Vidarbha, such as Nanded, Jalna, and Nagpur. The majority, however, belonged to closer regions such as Satara, Sangli, Kolhapur, Nasik, Solapur, and the Pune district itself. Many of them returned to their natal village to perform goddess rituals because, according to ritual diagnosis, the symptoms for their affliction had manifested only after marriage; this required them to return home to undo the original affliction knot, even if their repeated returns to far off natal villages were a constant bone of contention in their marital families.

Goddesses were even more mobile. Not only did they follow their devotees from their main shrine to the devotee’s village, crisscrossing the village’s own saṭvāī’s region and territory (Dandekar 2009), but the goddess also responded to her devotee’s worship across great spaces, as devotees only had to construct an impromptu shrine on the village outskirts, facing the direction of the main shrine, while performing the goddess’s rituals (this was mostly restricted to non-affliction rituals). I was told by an older woman (she was an afflicted woman’s mother-in-law) that though she was from a village near Yavatmal, and had never returned home in forty years since her marriage, the saṭvāī had never failed her, allowing her to have four sons successfully. She was disgusted that her daughter-in-law, on the other hand, needed to run home constantly because she was afflicted. But not all women were healthy enough to travel home. Women like Shruti, who were in the middle of reproductive crisis, were sent back to their natal families only as a last resort.

Shruti and Aparna shared their personal, very physical, and emotional family experiences with me, when Shruti was undergoing an intense phase of reproductive crisis, family breakdown, and saṭvāī diagnosis. These narratives also provide accounts of how they consciously negotiated with their village, marital clan, and caste during difficult times. I have treated their strategic negotiations as empowering, even if these may not formally fit classic definitions of feminism from a western perspective.3 As will become evident from the narrative, Shruti and Aparna were not merely my research subjects but had the power to understand, reflect on, and analyze their own situations. Neither do their narratives constitute a single interview, written out in one instance. I knew them both, their families, and husbands for a long period, and the narrative provided below is constructed out of a range of mutual interactions with them that were not always smooth. I have shielded their personal identities at the expense of sharing very little quantitative detail about their village, family statistics, or monetary status, since my data reveals how many afflicted and childless women, despite caste difference and status-disparity, share similarly oppressive diagnostic experiences once they are childless. Also, since I cannot provide multiple case studies to compare how social and economic contexts transform women’s micro-narratives of affliction, which in any case is quite outside the purview of my research, I have instead retained my focus on the narrative of Shruti and Aparna as conscious personal journeys of affliction, to foreground their crisis that resists inclusion into typical narratives of ritual healing.

The Narratives

Shruti and her family belonged to the Maratha caste (considered a dominant and upper caste in the village). When I first met her, she introduced herself to me as a graduate and told me that she was working before her marriage in a small rural bank, but had later left her job after marriage, when she had a child. At the time I had the impression that Shruti wanted to communicate important information about her background as a woman who was not just modern and educated, but who was intelligent and worldly-wise enough to understand that her commitment to family constituted a worthwhile and conscious priority for her. This was a common attitude that many younger village women projected at the outset of my research. I felt this to be a way in which they indicated their empowerment to me. They were asking me through these mechanisms to not treat them as hierarchically lower, just because they were part of larger rural clan-caste networks. As modern and educated women, they wanted to evaluate my own work and life decisions in return for my research about affliction goddess diagnosis.

I was to encounter Shruti’s affliction diagnosis later, after another village respondent passed on the information to me as gossip. When I went to Shruti’s home, she told me that she had developed saṭvāī affliction symptoms after she had conceived a child for the second time. She already had one somewhat mentally disabled son (though I am not certain whether he was ever medically diagnosed with mental disability). She had not wanted any more children immediately and therefore had had an intra-uterine device (IUD), known in India as a “copper-T,” inserted within her uterus free of cost at the government hospital, after much discussion and argument with her husband and in-laws. This device was ideally supposed to have functioned as a contraceptive device, but since she conceived again, the IUD had failed. Shruti reported that she “knew” that she became pregnant again and also “knew” about having miscarried thereafter, even though there was no medical confirmation of this.

Even as she became very ill, she was diagnosed with saṭvāī affliction by a ritualist living in the nearby town, who enjoyed a wide reputation in the region for being very busy, rich, and famous as a result of her healing talents. Though the ritualist hailed from the village itself, she had begun living in a nearby town that was closer to the highway, due to its accessibility to her clientele. She had the reputation of possessing an important list of clients hailing from Pune, Mumbai, and even Delhi. Though she visited the local and older saṭvāī village temple, most daily temple duties were performed by her younger sister-in-law during the time I was in the village, while she herself, as the main ritualist, was playing the instrumental part in building a bigger and modern goddess temple in the town. Consequently, she had also become powerful in the village, even though she could hardly spare much time for local patients who were not members of a few important clans (Shruti’s being one of them). Since the healer was of lower caste (dhor) with extended family still in the village, the upper castes said that she could not afford to disappoint the really important families, and she would have to serve them even without payment.4 The healer herself was compelled into obsequious behavior, which, I suspect, was intended to deflect upper-caste attention from her fame, money, and importance, so that her stature would not cause offense and expose her extended and economically weaker family living in the village to reprisal in the form of caste-violence. She kept repeating how Shruti’s family was her āī-bāp (mother-father) at every ritual diagnosis occasion and that she would never disappoint them. She would even “roll at their feet.”

According to Shruti, there had been continuous bleeding and abdominal pain ever since the copper-T had been inserted. Sudden copious and painful bleeding, which Shruti said was unbearable and felt like a miscarriage, took place one morning when she had gone to the fields to relieve herself soon after the copper-T insertion. The nurse and midwife at the dispensary could not say anything of diagnostic importance about the pregnancy and miscarriage but identified the bleeding and pain to be caused by a prolapsed copper-T obstructing Shruti’s cervix. The nurse removed the copper-T and advised Shruti to rest and generally sympathized with her condition, but also expressed amazement at an event as rare as copper-T expulsion. She asked Shruti to return after three months for a new IUD refit.

Even though many slightly older women in the neighborhood, who themselves had IUDs, by then had tried to reason with Shruti, her painful physical experiences of the IUD (lower abdominal pain and bleeding) and its expulsion were so unfathomable and traumatic for her that she could only compare her suffering to a miscarriage. But a miscarriage obviously involved other problematic questions that included sexual relationships. When she kept insisting that she had suffered a miscarriage and pregnancy, however, her mother-in-law told me and a group of other women that Shruti and her husband had been asked about their sexual relationship to ascertain whether a real pregnancy could have taken place at all. But while Shruti had displayed confusion and had been unable to answer any questions (about the sexual relationship with her husband), Shruti’s husband had denied having had sex with her. I was furthermore given the embarrassing task of interrogating Shruti on the matter of her sexual encounters “one final time,” since Shruti’s mother-in-law believed that Shruti, as an educated woman, would somehow confide extra details to another educated woman.

I tried to tell Shruti that whatever had happened to her was normal and was not the saṭvāī’s punishment; she would soon feel better. Moreover, if she had had the IUD insertion according to proper medical rules, within two to three days after menstruation, how could she have been pregnant and miscarried? But she remained adamant, with her doubts, fears, and her experience of how it had all “felt” different taking precedence in the conversation. As she began speaking about the divine existence of the child, and of the goddess having detected her sin of rejecting and resisting impregnation by using the IUD, I began feeling doubtful too. There was, after all, the small and marginal chance of ectopic pregnancy, despite the IUD. When I asked Shruti whether the baby she had miscarried belonged to her husband, she stared at me blankly, saying that the baby belonged to the goddess saṭvāī and that the goddess had taken it from her as a punishment for using the IUD. I could not bring myself to ask her any more pointed questions about sex thereafter, and I left her house feeling unable to recognize the old Shruti, who had questioned me so closely on my research.

Shruti’s husband had already denied having had any sexual relationship with her in the recent months before the IUD, and this had intensified suspicion around her, transforming her illness into impurity, immorality, and affliction. Shruti had expressed that she had had a reproductive experience, wherein she had physically felt pregnant after the IUD and felt that she had miscarried when the IUD prolapsed. Shruti also felt afflicted by the loneliness effected on her by her husband, when he abdicated responsibility and participation in her experiences of a failed IUD.

The affliction diagnosis was made thereafter. The place where the affliction was supposed to have taken place, namely the fields where Shruti had gone that morning when the “miscarriage” took place, conveniently faced the back wall of the saṭvāī shrine, even though at a considerable distance. The use of an IUD as “weaponry” (śāstra-prayog) against conceiving children was viewed as the reason for angering the goddess. Hence, the saṭvāī had to be ritually appeased. Once, upon meeting the healer on a six-seater ride from the village to her town near the highway, I asked whether the ritual intervention would really help Shruti. The healer made hand gestures that told me that it was only due to her ritual diagnosis that Shruti was alive at all. It was good that she was only ill, since it showed that the rituals had worked enough to stave off death. I thought the ritualist was incensed by my questions, because she invited me to her home to view the boils and lesions that she had developed on her body due to having absorbed Shruti’s illness. She began extracting various visiting cards from her blouse to demonstrate the range of her clientele while I was visiting her. I realized that though it was easy for me to be skeptical of ritualists, they themselves were equally powerless and oppressed as those who faced ceaseless caste violence, especially in states such as Maharashtra.5

Shruti’s husband, mother-in-law, and sisters-in-law blamed Shruti. They were angry with her for introducing saṭvāī affliction into their family as a stigmatized inheritance, so took the stance of being righteous victims who had lost a potential heir (kuladīpak) because of Shruti’s willful desire for an IUD. It was now said that the IUD had always been intended to facilitate immorality, but the goddess had revealed Shruti’s sins (pāp) and taught the family a lesson to never trust their daughters-in-law again and give in to their wishes. There were lots of discussions about the lost son, even though his whole existence was in question, and this plunged Shruti into deeper depression and illness. Since Shruti herself feared the saṭvāī’s retribution and believed in the ritualist’s affliction diagnosis, it became convenient for everyone in her clan not to return to the doctor for further medical diagnoses or confirmation. Taking on the mantle of primary victimhood, they all continued to blame Shruti.

With the pretext of bringing iron and calcium tablets for Shruti every week, I visited her regularly during this time, and it was then that her mother-in-law told me that her son (Shruti’s husband) was also depressed, grieving the loss of his “son,” due to which he had taken to drinking. I asked Shruti’s husband to accompany me to the dispensary for Shruti’s tablets the following week and asked him about his drinking. While looking at the pictures of various social leaders on the clinic’s waiting room wall (Gandhi, Nehru, Ambedkar, etc.), he tried to answer me by pointing out how all the country’s leaders would not have been born if their mothers had all been using IUDs, and how they would not have been born either if their mothers had conceived them outside of marriage. I realized that Shruti’s husband had taken her at her word, deciding that she had actually confessed to having a sexual liaison outside of marriage, which he thought had always been her actual aim when being fit for an IUD. The goddess had revealed her immorality to the village and punished her through a miscarriage, despite her IUD. But now the clan had been forced to bear the brunt of the victimization effected upon them by Shruti. It made no difference to them at all that she was ill.

Both Shruti’s affliction and illness progressed as she withdrew socially (due also to excessive hair loss). She was gradually unable to eat, digest her food, or sleep. She continued to suffer from gynecological problems and fainting spells and was soon bedridden. She also developed discolored patches on her face and neck, which were all diagnosed as a part of affliction as well. Shruti’s mother-in-law told me not to come to their house, since the affliction was at its peak and could also infect me if I had my periods. I was to learn later that Shruti’s father and uncle had come to fetch her, taking her back to her natal village.

When I asked about Shruti’s saṭvāī rituals, I was told that Shruti had been too ill to perform them. Her younger sister-in-law Aparna had to perform these rituals in her stead, since she ran the danger of inheriting Shruti’s affliction too. Shruti’s narrative therefore remains that of reproductive illness and family breakdown diagnosed as affliction, after which she suffered collapse and left the village. I did not see Shruti return to the village but heard much later from the village grapevine that she had joined her husband in Aurangabad; they had both left the village.

Aparna was Shruti’s younger sister-in-law and according to gossip, she had grown angry with Shruti for introducing the affliction stigma into their clan. It had now become extended to her own person and children. She was also angry with Shruti for having initiated an environment of suspicion and distrust against the daughters-in-law in the family, wherein every demand made by Aparna after Shruti’s illness-affliction was first viewed as immoral and examined for its potential danger to the clan’s reputation by her in-laws. Every discussion about Aparna’s life had grown to become a comparison with Shruti’s, and she had also been instructed to perform Shruti’s rituals along with her own preventive ones in order to protect from the goddess’s affliction being visited upon the clan children. When I met Aparna at a mutual friend’s house, she was reflective and analytical about Shruti’s situation from the very beginning, which I believed was a conscious effort on her part to separate herself from Shruti and the way the latter had collapsed. Aparna was making it amply clear to me that she was not afflicted and that she had an “academic” opinion, even if she was not educated and modern like Shruti. She was also resentful about being identified with Shruti’s affliction by me and recounted how she had increasingly begun to answer solicitous or curious questions about Shruti with angry answers of “I don’t know!”

Performing saṭvāī rituals (Aparna had not started with the rituals yet, when we were talking, but had received instructions and directions) would require Shruti (or in this case, Aparna on her behalf) to accept the ritualist’s affliction diagnosis at the goddess’s shrine located outside the village. This entailed apologizing for Shruti’s mistake/misconduct that had resulted in transgressions, and would prevent the mistake occurring in it in the future for Aparna as well. After this, food and offerings of fertility (bangles, vermillion, turmeric, a comb, flowers, and a blouse-piece) would be made repeatedly at the shrine every week for five weeks. After every sojourn to the shrine to see the chosen representatives of the goddess—five virgin pre-pubescent girls, an old widow, and a saṭvāī ritualist or a midwife—ritual food would be consumed at home. Women without menstruation, women without reproductive capacity, and women who perform cleaning tasks occupationally are considered immune to the saṭvāī’s affliction and are therefore culturally considered her agents.6 The ritual at the shrine is usually witnessed by the entire network of the afflicted woman’s clan in the village, constituting her caste along with the ritualist.

The most important aspect of saṭvāī rituals was the undertaking of a strict fast by Aparna (on behalf of herself and Shruti), wherein she would abstain from all the foods and ritual articles that were offered to the goddess during the five-week ritual period (consisting of bread, sweets, rice, fries, certain vegetables, milk products, and rich foods). This fast acknowledged that women returned to the saṭvāī a claim over their food and fertility as something that was not rightfully theirs. In so doing, they marked their motherhood duty as part of a bargain that they had not fulfilled (or were fearful of leaving unfulfilled) in return for their claims to be wives in marital clans, with the accompanying rights to food and gifts. Shruti (and Aparna on her behalf) was forbidden to eat outside her home during the ritual period and hence declined all invitations lest she felt tempted to eat foods that she was supposed to ritually renounce to the goddess. This also protected her from eating at the hands of secretly menstruating women, since there was always the fear of contracting contagion. Aparna was hence publicly ritually quarantined due to Shruti’s “faults.” People said that she felt angry about the public quarantine.

To my surprise, Aparna’s opinions about Shruti’s affliction, on the other hand, remained reflective, pragmatic, and sympathetic, yet analytical. I was a little heartened to discover that Aparna was not angry with me for having helped Shruti and that she too considered herself to be ultimately in Shruti’s camp, since they were both co-daughters-in-law. She considered the extending of affliction to herself irritating, however. She said that Shruti had been impulsive and foolish (mūrkhapaṇā) to carry on in the miscarriage vein, when they, Shruti and dā-jī (Aparna’s brother-in law/Shruti’s husband) had obviously had no sexual contact. According to Aparna, it was this that had antagonized dā-jī considerably and left the clan with no other option but affliction diagnosis, since Shruti herself believed in it. Shruti’s own belief in turn had led to the intense gossip. Emboldened by Aparna’s preparedness to analyze Shruti’s situation sympathetically along with the small steps she herself was taking in everyday life to negotiate with her marital clan’s sudden heavy-handedness with its daughters-in-law, I asked Aparna her opinion on Shruti’s IUD decision and the ritualist’s verdict of the saṭvāī affliction.

Aparna said that Shruti had been hasty about the IUD decision because she had only one son, who was, moreover, mentally challenged. Aparna felt that Shruti could have waited and allowed the birth of another one or two children before taking the contraceptive step. After all, many village women had a copper-T, and IUDs were free at the hospital. Even if subsequent children had been daughters, the clan would have surely given in to a contraceptive. According to Aparna, Shruti should have satisfied the clan and her husband’s expectations of a kuladīpak first before drawing attention to her own demands so openly. Once they were satisfied, she could have deflected attention from herself and followed her own wishes more quietly. Aparna went on to tell me how many people in the family and neighborhood had advised Shruti along similar lines: to delay the contraceptive decision, since she had only one child and was as yet only in her mid-twenties. But Shruti had remained adamant, and it had become a matter of winning an argument against the family for her. So she won in the end and the IUD was fitted. I asked Aparna the reason for Shruti’s sudden capitulation after she got the IUD from a state of excitement about winning a family argument into narratives of punishment, saṭvāī affliction, and miscarriage, even though she knew that there had been no sexual contact between herself and her husband. Aparna did not know, but said that she could hazard a guess about Shruti’s tragedy, based on what she knew about her.

According to Aparna, the fact that IUDs were not always comfortable was well-known, and although many women who had copper-Ts themselves tried to comfort Shruti with information about their own uncomfortable reproductive health experiences, this did not help Shruti because the side-effects in Shruti’s case were probably much worse, unbearable, and unexpected. Moreover, getting the IUD was associated with Shruti’s victory. The fact that it disagreed so vehemently with her might have disappointed Shruti more than others understood. This could have made Shruti feel that she had done something wrong and was punished for it. Aparna pointed to how both affliction and illness came together for Shruti, who perhaps hid behind an affliction diagnosis, since she was perhaps too proud to accept her mistake and get the IUD removed as many women do. Later, she hid behind illness to escape her husband’s and clan’s anger about the immorality accusations that always accompany saṭvāī affliction diagnosis. Finally stuck between both illness and affliction, she collapsed and was sent back to her father.

Aparna’s opinion of the ritualist remained caustic and negative. According to her, the ritualist’s riches in the nearby town did not change her lower caste and vulnerable status in the village. In fact, it made her more vulnerable because she had to counterbalance her individual power with village upper castes even more carefully, since she had more to lose by growing rich. Members from upper castes could confiscate her money whenever she displeased them. The ritualist would have to forever remember that her capital, riches, and fame rested on her own caste occupation that cleaned reproductive impurity and affliction from upper-caste women.

An Overview of Saṭvāī Affliction and Diagnosis

What I understood with nearly every saṭvāī diagnosis case that I encountered was that it was not afflicted women but their marital clans that were ritually recognized as the “real” victims in cases of child-loss. Mothers were held culpable and guilty for having caused these afflictions to their clans, even when they fell very ill. Moreover, saṭvāī ritual diagnoses were not just judgments that indicted individual women who were already suffering reproductive crisis, but a systemic and patriarchal diagnosis that established the clan as the primary, innocent victims. Since the clan’s body politic in the village had the status of insiders, they required and deserved healing that restored their rights to an heir from the goddess. In many cases, wherein afflicted (and sometimes very ill) women were unable to conceive children, even after performing rituals, they were sent back to their natal families as failures, while their marital clans retained a wounded and righteous pride about their own innocence. From the perspective of the saṭvāī paradigm, therefore, afflicted women suffering child-loss were not primary victims but secondary and ambivalent victims-cum-perpetrators who needed to be exorcised in order to heal primary victims (i.e., the marital clan), so that the “othering” and evil illness besieging their clans and their fertility could be removed.

Afflicted women were therefore viewed as potential others and a weak link in the marital clan’s fertility, and it was therefore the afflicted woman who had to repeatedly inhabit the boundary spaces of the goddess’s shrine (that was always located outside the village boundary) in the saṭvāī’s rituals, while making ritual offerings in order to publicly demonstrate the danger she constituted to her marital clan in the village. It was she, when suffering child-loss and unable to provide her marital clan with children, who had to bear accusations of either being associated with or influenced by a witch7 or herself being a “child-eating witch,” even if her own child had died. When and if these women were ultimately able to have sons, a history of child-loss, if diagnosed as saṭvāī affliction, would continue to stigmatize them as women who possess dangerous witchcraft potential. This required their sisters-in-law, daughters, and daughters-in-law to follow saṭvāī rituals in a preventive form in order to appease the goddess and evade inheriting the affliction.

This placement of women as ambivalent, secondary, and guilty victims served to divest them from claiming their reproduction, motherhood, and children as their own. Instead, the birth of children, especially sons, was viewed as beneficial for a woman’s marital clan inclusion, since and the birth their legitimacy of was considered a form of “purity” children, which that would ultimately contribute to their children’s longevity and the woman’s valor as an auspicious mother. Children were rightfully to belong to their fathers, patriarchal clans, and village castes. Childbirth and motherhood were hence viewed as a woman’s marital duty and labor provided to clans, since child-loss was blamed on them. Those who failed their motherhood duty, therefore, faced a severe breakdown in their social relationships and emotional well-being. Consequently, they had to confront physical illness, neglect, and sometimes domestic abuse.

The saṭvāī’s ritual paradigm of affliction therefore assumed larger and more complex proportions than just a form of healing evil illnesses through exorcism. It progressed into a form of bodily discipline and fear among women, wherein evading punishment became a normative way of leading life under ritualized forms of law.8 Child-loss within this context prevented women from grieving9 and presented them instead as people facing public disgrace, culpable, and subaltern among other women, especially among auspicious mothers of sons, who occupied an upper social strata in rural communities that spanned across caste. Mothers of sons in my field area, who claimed to be fertile, auspicious, and to have never been associated with child mortality, stigmatized and discriminated against childless women or those mothers who had suffered child-loss in the past and had been ritually diagnosed as afflicted by their families. This stigma remained even if these afflicted women had been later successful in having children of their own. The stigma of being marked by child-loss continued as a sign of being inauspicious for many women, who, like Aparna, had to continue performing preventive rituals periodically but were also increasingly excluded from social invitations due to shared affliction diagnosis. Two categories of women, viz. mothers and witches or those subjected to witchcraft, were therefore produced across caste in many rural areas where the saṭvāī held sway. The boundary between women who were condemned as witches and women who claimed bewitchment was often so blurred that women claiming victimhood and bewitchment, who sought healing and clan reintegration, often turned against or pitied those who were caught out by the goddess and condemned as immoral or as having witch-like potential. All afflicted women were, however, stigmatized through mechanisms of social ostracism that included not being invited to social occasions, celebrations, and festivities in the village. Their food utensils were separated, destroyed, or given away, even if they were from important families in the village and had to be invited.

Apffel-Marglin 2008 describes how the smallpox goddess in Orissa induced disease as a form of discipline among her worshippers. This motif also becomes very popular in the worship of other disease goddesses (described in the maṅgalkāvyas by scholars such as Vanashree 2010) that act as village deities (described in South India by scholars such as Brubaker 1978 and Masilamani-Meyer 2004). A gendered analysis demonstrates the importance of moral discipline especially among women and subalterns, as upholders of dharma (duty). Women and subalterns are marked as guilty and culpable of the crime of non-adherence and immorality, especially when those suffering its repercussions are identified as innocent sufferers and are proven to be victims. Many women in my field area referred to their motherhood duty as dhamra. I first thought that dhamra was an instance of mispronunciation, but when I noticed that dhamra co-existed with dharma, I began questioning my respondents. Though articulation of conscious definitions remains an academic preoccupation, many among my respondents explained the meaning of dhamra to me with the help of body language that involved first pointing in the direction of the saṭvāī shrine and then shaking their fists in the air to indicate her punishment and discipline.

Since saṭvāī afflictions are complex, with layers of primary victims and ambivalent, secondary afflicted-perpetrator layers, saṭvāī ritualists play a complex role in diagnosing and healing as well, since their priority and loyalty lies with upper caste clans in the village. Research on childbirth rituals and birth attendants (Chawla 1994; Chawla 2002) has already pointed to the perceived impurity of midwifery, due to its association with women’s menstrual fluids and childbirth-related vaginal secretions that are also considered impure and dangerous in beliefs prevalent across South Asia (Bennett 2002; Raheja and Gold 1994). Midwives are therefore linked to the management of what is culturally understood as female forms of danger, since menstrual fluids are simultaneously considered to be extremely potent when deployed within witchcraft (especially by my respondents). Therefore, the management and cleaning of these fluids by midwives made many midwives ritual agents of upper-caste village clans in my field area, in what many of them expressed to me was a form of caste-based balutedāri (occupational tasks such as pottery, weaving, etc. performed for upper castes, especially on ritual occasions, in exchange for payment in kind and patronage, similar to the jajmāni system).10 Clans retained a primary interest in cleaning female danger and witchcraft-inducing impurity that might cause their women affliction and them victimization through child-loss as well, and they therefore patronized lower-caste ritualists from the perspective of being the primary victims of afflicted women. Many midwives acted as saṭvāī ritualists in my field area too, though not all saṭvāī ritualists were midwives. Midwifery, as Sadgopal 2009 has amply demonstrated, is also a skill.

Even as the healing rituals of upper-caste afflicted women necessitate physical contact with lower-caste ritualists, this association in my field area is completely controlled by the afflicted woman’s clan members. All conversations at ritualized moments between afflicted women and saṭvāī ritualists are heavily monitored, remain limited, and are hardly subject to the ritualist’s own terms. The ritualist is immediately made aware of her service-provider station if she tries to counter the clan’s narrative. In saṭvāī worship, therefore, the primary interaction for afflicted women at the shrine does not take place between her and the saṭvāī ritualists at all, but between her in her own afflicted position, and the saṭvāī, via and finally the prescribed ritual that charts her personal journey from being afflicted to making ritual offerings that signify her possessive motherhood to the goddess.

The saṭvāī always recruits her own ritualists (often inter-generationally and often by sending miraculous dreams) from among Hindu lower-caste women (mostly of the mātang caste). The saṭvāī provides (“blesses”) them with the auspicious valor of motherhood and the ritual power to clean the contaminating affliction from upper-caste women’s bodies. Since rituals that rid women of reproductive contamination (known as viṭāḷ) are performed by afflicted women at saṭvāī shrines, ritualists only perform a diagnosis of affliction and describe its nature and the location of its occurrence to the afflicted woman’s clan members.11 They further attend the ritual performances made by afflicted women at the shrine and at the home of the afflicted to accept food offerings and gifts. Their ritualized cleaning services are therefore specifically characterized only by their physical presence at shrines and at rituals, not by any specific healing rituals undertaken.

When I asked my respondents about the role saṭvāī ritualists play in the healing of afflicted women, they told me of how their lower-caste presence was of utmost importance at rituals, since the actual ritual was performed only by the afflicted woman or by mothers who wanted to prevent affliction. The ritual was supposed to open a door or a communicative avenue or page (pān) between afflicted women and the goddess, and it was through this door that already afflicted and potentially afflictable women returned “whatever was not rightfully theirs” (i.e. their fertility, thereby counteracting possessive motherhood) to the goddess. Women in preventive rituals always laid their newborns in front of the saṭvāī as well. During such dedications, women’s affliction and viṭāḷ (even if withheld secretly) seeped out of her, positing a threat to other pure clanswomen, who were her witnesses, for this vow. The lower-caste ritualist’s body, therefore, had to be present at the ritual to absorb this seeping viṭāḷ into her own body and thereby prevent affliction and contagion to other upper-caste clanswomen present at the ritual. In return for this, she was made gifts and included as a recipient for ritual offerings. Saṭvāī ritualists became important among women of upper-caste clans as they were often invited for a meal when a child fell ill, so that her lower-caste absorbing presence, rewarded by food, could help to absorb the child’s mother’s own viṭāḷ and rule it out as a reason for the affliction causing the child’s illness. When I asked my respondents about why lower-caste ritualists were never afflicted by viṭāḷ themselves, I was told that they were immune to any upper-caste impurity and affliction, reproductive or otherwise, since undertaking cleaning operations was part of their caste-based occupation and balutedāri, just as motherhood was women’s. Indeed, those whom the saṭvāī had specially chosen definitely encountered affliction, according to my respondents, in cases where they ignored the saṭvāī’s dream-commands to become ritualists at shrines or if they disrespected their balutedāri, just like married women became afflicted if they disrespected their balutedāri of motherhood.

But there were obvious problems here. While some saṭvāī ritualists I spoke with did consider themselves lucky to enjoy village patronage and the goddess’s blessings despite being poor and of low caste, almost all of them spoke of reproductive illnesses (such as menstrual cramps that kept them in bed, a swelling ball-like feeling in their stomach while urinating and defecating during menses, white and red discharge, blood-loss, and exhaustion) they had contracted out of the physical absorption of viṭāḷ from upper-caste female client bodies, almost as an occupational hazard. Any suggestions of seeking medical aid made them retort with exclamations of how the goddess would curse and afflict them with child-loss if they did so, and of how these illnesses were part of their job and the payment in return for enjoying the goddess’s blessings.

Trance was a difficult subject too. Although most of them admitted to beginning their journey as saṭvāī ritualists with an ability to diagnose afflicted women in trances (and I attended many ritualist trances during fieldwork myself), many said that trances troubled them later, making them ill (trās hōtō). So, they used divine powers from trances gained by them in the initial stages of their jobs as saṭvāī ritualists to continue with their diagnostic work later. There were obvious problems with the balutedāri of motherhood as well, because even after having children, mothers had to continue with preventive rituals, since the reproductive impurity of their viṭāḷ continued, which would socially produce them under affliction diagnosis as witch-like child-eaters, even after their motherhood duties were fulfilled.

In terms of caste, therefore, the subjectivity of being afflicted by the saṭvāī becomes revealed as one wherein women, who are defined by caste, lose their caste purity, which in turn results in child-loss. This is rather significant, since the child, who would have otherwise been of pure caste, alive, and part of its father’s clan if its mother’s afflicting contamination were absent, implies the nature of the mother’s viṭāḷ to be of a sexualized variety. The contamination of caste through reproduction that produces the mother as the sexualized and immoral other becomes reminiscent of many village goddess narratives in the Deccan and Maharashtra, such as Mari-ai, Renuka-ai, and Yellamma (Hiltebeitel 1988).

Even the unintentional sexual indiscretion of upper-caste women is a familiar story, which forms the basis for their condemnation as impure, the dismemberment of their motherhood from their own selves, and the conjoining of their bodies with lower-caste women to ward away excommunication by being redeemed both as goddesses and by goddesses. The making of ritual offerings to the saṭvāī encompassing “what does not rightfully belong to women” also confirms the conceptual dialectic observed by Doniger 2009 that differentiates the passive female fertility and the active masculine sexuality in literature. Women seek a restoration to their passive and fertile feminine form, which would help them in their motherhood duty, even as their viṭāḷ is cleaned by ritualists.

Can Narratives of Affliction Replace Ritual Healing?

The final discussion can be divided into three topics. The first relates to the different meanings that could be read in Shruti’s IUD expulsion in terms of critiquing the Indian state’s medical contraceptive policies. The second topic that is interrelated with the first explores how women’s narratives of affliction can form a separate category within discourses of ritual healing. The third concerns hierarchies and associations formed by conflations between gender, caste, and clan, which lead to the debacle of the lost child.

Supporters of tradition as healing may prefer me to read Shruti’s affliction narrative in other ways, since alternate readings are always possible. Alternate readings of Shruti’s story would focus analysis on her body’s expulsion of the IUD as an almost somatic rejection of modernity, as Shruti chose to believe in the traditional diagnosis of saṭvāī affliction. But such readings would also reduce Shruti’s narrative, forcing it to fit into predetermined structures, glorifying tradition as healing rather than letting it remain what it was: an affliction narrative. Though I knew Shruti was ill and was diagnosed as afflicted, I still do not know whether she was ever healed. Besides, reading her story thus would selectively obliterate a very important aspect of Shruti’s self-projected identity: a woman who prided herself as modern and educated, who chose to assume control over her reproductive body against her marital clan’s wishes, who participated in government programs for rural women’s reproductive health, who once worked in a bank, and finally, who used all these attributes as empowerment, someone who consciously chose to settle into family life and motherhood. Shruti’s narrative was obviously more complex and layered as within the affliction diagnosis her marital clan received the primary position of victimhood (in contrast to her herself), and in the process, her independence as a modern woman who argued for her own will failed. Her narrative demonstrates her secondary and ambivalent position as victim-cum-perpetrator, which is a classic situation within saṭvāī afflictions. Neither does the saṭvāī’s ritualist posit any traditional or alternative knowledge for scholars apart from an understanding of how lower-caste bodies are used as labor, when ritually introduced into upper-caste clans, to clean their women of viṭāḷ.

As far as the second aspect is concerned, the postcolonial praise of tradition as healing (Kakar 1982; Nandy 1989; Sax 2010) posits intensely written narratives of rituals that heal, rituals that are performative and educational, including ritualists who go into trance, ritual texts, different techniques, strategies, processes, and contextual relationships. In academic descriptions, ritual healing and efficacy is theoretically located critically at the juncture of a failed, colonialist, and inadequate modernity, where it develops lacunae. Redemption is also scripted in thickly described collective spaces and the intricacy of social relationships in accompaniment with comforting images of deities and traditions. Narratives about ritual healing are written, however, almost entirely at the expense of individual women’s narratives of oppression and negotiation, while many afflictions that cause the family to take center-stage as primary victims of women’s afflictions are gifted by the same deity, who is constructed by these narratives as a healing agent. It is only closer examination of individual case studies that reveals afflictions to be a source of patriarchal discipline and punishment for married women. Afflictions serve to place women in a position secondary to patriarchal clans. Discourses of ritual as healing almost always locate redemption within traditional collectives of clan, caste, or village, collectives which assume consensus and a shared and linked subjectivity, using paradigms of “dividuality” ( Marriott 1990) as being the primary cognitive and experiential framework for South Asians. This limits discursive opportunities for understanding women’s affliction that describe selfhood within contexts of excommunication, wherein individual experiences of an ambivalent boundary are generated. Women’s sharing is often a conscious choice, based on their individual sexual, reproductive, and clan relationships. Their negotiations are personal and often strategic.

It is often expected that affliction narratives will tell only half the story. It is almost always taken for granted that the entire narrative will only be complete when affliction is a precursor to ritual healing. Kakar’s book on India’s healing traditions (1982), for example, remains influential among many South Asian social scientists, psychologists, medical doctors, and psychiatrists. It demonstrates examples of various afflictions and maladies in ways that showcase the power of Indian mysticism in being capable of healing patients through magico-religious techniques, which become metaphorical for demonstrating the power of tradition to heal the illnesses generated by modernity itself. Affliction is thus utilized only to tell the healing story, which unfolds in the latter part of a case-story to engulf affliction and obliterate it. It is expected for ritualists to engage with the afflicted to complete the healing, and culturally contain the m affliction in ways that are alternative to the methods of modernity. These exchanges are then thickly described in different ethnographic contexts to prove a certain point as scientific: that ritual is universally healing.

Although this might indeed be true even of many cases, ritual healing does not fit women’s affliction narratives of reproductive health and childbirth crisis so easily, when most of them are diagnosed as witches. Saṭvāī affliction narratives are not success stories. Neither are these stories ever really complete, nor do women ever become totally healed or receive enough closure, since afflicted women continue to remain stigmatized as impure once accused of witchcraft and immorality. They remain hovering around the periphery of their marital clans, who are established as the primary victims in their diagnosis. Afflicted women do realize that ritual diagnosis is a form of punishment and suffering, since they also live within modernity, however fragmented its institutional execution may be. Even if they are able to have children, they remain vulnerable to a fear of affliction diagnosis, due to their personal history with it. Neither do the ritualists play too important a role, because they serve marital clans as well. My research on afflicted women as exorcized elements, condemned as impure and sometimes as child-eaters, could be viewed as narratives of ritual affliction that therefore resist being subsumed into the category of ritual healing.

Arriving at the last point of my discussion, gender and caste in Aparna’s narrative is conflated as she speaks of clan negotiations. She denigrates the ritualist for the same strategy she recommends Shruti to use. This seems contradictory. In Aparna’s estimation, Shruti should have first fulfilled her husband’s and marital clan’s reproductive demands in order to deflect attention from herself to later fulfill her own contraceptive agenda (a step-by-step negotiation that Aparna was herself undertaking). It was this strategy that the lower-caste ritualist was also employing. The ritualist had built and earned her own fame and money, but had always fulfilled the wishes of upper-caste clans in the village, too, remaining careful to always prioritize while never antagonizing them. While this strategy for Shruti, in Aparna’s opinion, would have gained the former freedom, empowerment, and respect, she viewed the same strategy that the ritualist employed as a cunning ploy to subvert her low-caste station. Aparna further pointed to how Shruti would gain clan support and motherhood with patience and strategic planning, but she also pointed out how the same strategic planning and patience left the ritualist exposed, at the mercy of upper-caste clans, as the ritualist had more to lose now and could be attacked by upper castes at any time.

First, Aparna was obviously demonstrating a hierarchy between two forms of motherhood-related balutedāri by the discrepancies in her attitude toward Shruti and the ritualist, wherein the actual provision of motherhood to upper-caste clans through marriage defines a higher rank for women in comparison to the ritual aiding of upper-caste motherhood by the absorption of viṭāḷ. Second, what Aparna was expressing more vitally was a competitive desire for power over her clan in comparison to the ritualist, which she felt went missing in Shruti’s case, only because the latter had antagonized her husband. Aparna was constructing two hierarchical rungs of power around her clan through her narrative. While she was intensifying the inner ring of mothers (Shruti and herself) as its strong wives and mothers who could negotiate with the clan with patience and strategy, she was competing with the outer rung (traditional service-providers of rural clans such as ritualists) by trying to distance them from the clan. More important than knowing whether Aparna was casteist was to understand that she was using caste discrimination as a means of controlling her marital clan and husband.

The question I often posed to my respondents during Shruti’s illness was this: whose child was it? Shruti said that the child she miscarried belonged to the saṭvāī. If this was the case, how was it that her marital clan claimed victimhood for her miscarriage? Since her husband had taken no responsibility for being the father, how was Shruti’s marital clan wounded? How could Shruti’s miscarriage be discussed by them as their bereavement? If she had been accused of immorality (and then the child had no father), didn’t this make it her own child? And hadn’t the saṭvāī taken the child away, precisely because a fatherless child was disallowed from living in a morally patriarchal society according to dhamra? Even if Shruti was guilty of immorality, why could she not grieve the death of her own (even if illegitimate) child? This question was never really answered.

I asked Aparna this question too. Her answer was equally non-specific. The mutual friend at whose house we met to discuss Shruti was to concluded the discussion by rebuking me. She said I asked too many “why” (kāran) questions, when I actually knew the answer for these, in my heart. Since I could not answer questions myself, however, I wanted them to spell things out for me. She said with finality that marital families owned the bodies of women so that they could do whatever they liked with them, even kill and abandon them. They were thus offended with everything that these owned bodies did according to their own free will. The dead babies were only a way of exposing and shaming the free will (svēcchā) of women, who existed within bodies that were owned. There was no dead baby!


  1. The story of ancestral women has also been noted by Rairkar 2007 in her research with midwives in rural Maharashtra.  

  2. Feldhaus 2003 and Dandekar 2009 each document how both the saṭvāī and Janai goddesses followed their devotees to their villages, on the promise that their devotees wouldn’t turn to look at them. They turned to stone at the very place their devotees broke their promise and turned to look, which was usually at the village boundary.  

  3. Davis 2014 research on Maithili women’s stories is an excellent example of women’s small and meaningful victories, even when they are often pitted against each other, in negotiating complex and moral battles, interwoven together in narratives. Also Vanashree 2010research on Mahasweta Devi’s play titled Bayen, concerning witchcraft accusation, analyzes the text within postcolonial feminist frameworks that present subaltern women as resistant and empowered.  

  4. Dhor is officially supposed to be of a tanner jāti (caste), administratively considered a scheduled caste in Maharashtra. The family of the healer in question, still living in the village, said that they were of the Valmiki caste, which is synonymous with Dalit. Many from the Valmiki caste followed the Ambedkar movement and had joined neo-Buddhism. Some, however, like the healer’s family, were as yet Hindu and performed their caste duties, such as goddess healing and midwifery that were considered impure, since they also earned from it. When I asked about neo-Buddhist conversion plans for the future, they seemed optimistic, saying that they would consider conversion if the future generation secured good jobs and caste reservation. Till then, they were dependent on ritual cleaning tasks that paid more than the actual cleaning of gutters, streets, or toilets.  

  5. Quack 2012 has conducted a detailed ethnography of the anti-superstition movement in India, and located the movement within a trajectory of debates about modernity. 

  6. Chakravarti 1995, in her research on widowhood in South Asia, demonstrates how Brahminical prescriptions of sexually “castrating” widows include injunctions for them to avoid the “anomaly” of female asceticism, since women are considered as inherently lascivious. While these prescriptions lead systemically to sati (widow-burning), Chakravarti’s description of Havik Brahmin widows from South India reveals the social condemnations and stigma they face, wherein widows are held responsible for the death of their husbands, receive no sympathy, and are accused of cannibalism or eating up their husbands. I suggest there is a deeper stratum of witchcraft accusation embedded within these accusations. I also suggest that all saṭvāī ritual recipients represent systemic anomalies in the context of ritual feasts: midwives and ritualists, who are auspicious despite being impure, and prepubescent girls. The latter are an anomaly because they are legitimate clan children, without being boys and heirs of their patriarchal-clan. Because they are pre-menstrual, they cannot be wives and mothers to begin their destined function of clan balutedāri (occupational duty).  

  7. Although general anthropological literature on sorcery and magic exists (cf. e.g. Nabokov 2000), very few accounts on witchcraft allegations and witch-hunting are available, apart from examples that analyze and document Adivasi regions. To name a few, Chaudhuri 2012 research explores witch-hunts among the Adivasi tea-plantation workers of West Bengal; Mullick 2000 research explores witch-hunts in Jharkhand; Skaria 1997 research explores witchcraft accusations among the Dangs of Western India, while Sundar 2001 analyzes witchcraft accusations in Bastar. While this creates disputed opinions about witchcraft, such as it resulting from superstition, or witch-hunts being an Adivasi problem, or witchcraft being a product of poverty or a lack of education, it also demonstrates an absence of focus on the part of the researching self that leaves upper-caste society unmarked. The theoretical understanding of witchcraft emerging from such research is also borrowed from western theories that are better suited to understanding witch-hunts in Europe or America: eliminating socially dominant women by accusing them of witchcraft to claim their property. This hardly fits South Asia. Research on witchcraft accusations based on saṭvāī afflictions firstly demonstrates that witchcraft is definitely expressed as popular practice among dominant, upper castes, as well as among educated members of society, who are completely aware of debates about rationalism and superstition, live within a modernity, and are far from being poor. Neither are accused and afflicted women more or less dominant than other women, they are not eliminated, and no-one’s property is grabbed. Women are themselves owned. Desai 2008 ;Desai 2009 research , even though set among the Gond in Chhattisgarh and Maharashtra, is more nuanced, dealing with questions of Hindu nationalism, witchcraft allegations as ways of understanding Maoist insurgency, intra-community friction leading to witchcraft allegations, subaltern vegetarianism, and the adoption of Hindu deities for healing, such as mahānubhāv saints.  

  8. One of the most interesting facets of Foucault 1995 of relevance to this study was his theorization of punishment that transforms punishment into a culture, wherein its prevention turns into normative discipline, culture, and tradition. Connerton 1999 further adds to this theory by pointing to how these historical cultures of law are memorized within the body discursively.  

  9. Scheper-Hughes 1992 points to cultures of motherhood, contextualized within the poverty and slums of the Alto, Brazil, where women become disinvested from motherhood as soon as their children suffer malnutrition and become sickly. But there is also a stark difference between the women of Brazil, described by Scheper-Hughes, and the upper-caste women of Maharashtra, since the latter suffer impoverishment due to motherhood roles and duties imposed on them within patriarchal systems, and not malnutrition and poverty.  

  10. There is not much specific research on balutedāri as a system of interaction between jātis that ran parallel to the jajmāni system, in early modern Maharashtra, apart from Anand 2005. Balutedārs have now become organized according to a new awareness of four castes that are linked to reservation. 

  11. I have described saṭvāī rituals in detail elsewhere, for example in Dandekar 2016


  • Anand, Mily Roy. 2005. The State and the Village Community in Medieval Maharashtra (Seventeenth–Eighteenth Century AD). New Delhi. Rajat Publications
  • Apffel-Marglin, Frèdèrique. 2008. Rhythms of Life, Enacting the World with the Goddesses of Orissa. New Delhi. Oxford University Press
  • Bennett, Lynn. 2002. Dangerous Wives and Sacred Sisters: Social and Symbolic Roles of High-Caste Women in Nepal. New York. Columbia University Press and Kathmandu: Mandala Book Point
  • Brubaker, Richard Lee. 1978. The Ambivalent Mistress: A Study of South Indian Village Goddesses and Their Religious Meaning. University of Chicago
  • Chakravarti, Uma. 1995. Gender, caste and labour: Ideological and material structure of widowhood. 30: 2248-2256 Economic and Political Weekly
  • Chaudhuri, Soma. 2012. Women as easy scapegoats: Witchcraft accusations and women as targets in tea plantations of India. 18: 1213-1234 Violence against Women 10.1177/1077801212465155
  • Chawla, Janet. 1994. Child-Bearing and Culture: Women Centred Revisioning of the Traditional Midwife; The Dai as a Ritual Practitioner. New Delhi. Indian Social Institute
  • Chawla, Janet. 2002. Hawa, gola and mother-in-law's big toe: On understanding dais’ imagery of the female body. 147-162 The Daughters of Hāritī: Childbirth and Female Healers in South and Southeast Asia edited by Santi Rozario and Geoffrey Samuel, London. Routledge.
  • Connerton, Paul. 1999. How Societies Remember. Cambridge. Cambridge University Press 10.1017/CBO9780511628061
  • Dandekar, Deepra, and Abhijit Dandekar. 2011. The Satvai and settlement pattern in rural western Maharashtra. 27: 221-224 South Asian Studies 10.1080/02666030.2011.614430
  • Dandekar, Deepra. 2009. Satvai and the lives of women in Ghodegaon. 281-292 Liebe, Sexualität, Ehe und Partnerschaft. Paradigmen im Wandel: Beiträge zur orientalistischen Gender-Forschung edited by M. R. Roswitha Badry and Karin Steiner, Freiburg. Foerdergemeinschaft wissenschaftlicher Publikationen von Frauen
  • Dandekar, Deepra. 2014. Childlessness and empathetic relationships. 14: 123-139 The Oriental Anthropologist
  • Dandekar, Deepra. 2016. Boundaries and Motherhood: Ritual and Reproduction in Rural Maharashtra. New Delhi. Zubaan Books
  • Davis, Coralynn V. 2014. Maithil Women’s Tales: Storytelling on the Nepal-India Border. Urbana-Champaign. University of Illinois Press 10.5406/illinois/9780252038426.001.0001
  • Desai, Amit. 2008. Subaltern vegetarianism: Witchcraft, embodiment and sociality in central India. 31: 96-117 South Asia: Journal of South Asian Studies 10.1080/00856400701874734
  • Desai, Amit. 2009. Anti-‘anti-witchcraft’ and the Maoist insurgency in rural Maharashtra, India. 33: 423-439 Dialectical Anthropology 10.1007/s10624-009-9135-4
  • Dhere, Ramchandra Chintaman, and Anne Feldhaus. 2011. Rise of a Folk God: Vitthal of Pandharpur. New York. Oxford University Press 10.1093/acprof:oso/9780199777594.001.0001
  • Doniger, Wendy. 2009. The Hindus: An Alternative History. New York. Penguin Press.
  • Feldhaus, Anne. 1995. Water and Womanhood: Religious Meanings of Rivers in Maharashtra. New York. Oxford University Press.
  • Feldhaus, Anne. 1995. Types of river goddesses in Maharashtra. 31-47 Folk Cultures, Folk Religion and Oral Traditions as a Component in Maharashtrian Culture edited by Günther-Dietz Sontheimer, New Delhi. Manohar
  • Feldhaus, Anne. 2003. Connected Places: Region, Pilgrimage, and Geographical Imagination of India. New York. Palgrave Macmillan.
  • Foucault, Michel. 1995. Discipline and Punish. New York. Vintage Books
  • Gordon-White, David. 1996. The Alchemical Body: Siddha Traditions in Medieval India. Chicago. The University of Chicago Press
  • Hiltebeitel, Alf. 1988. The Cult of Draupadi 1: Mythologies: From Gingee to Kurukshetra. New Delhi. Motilal Banarsidass
  • Kakar, Sudhir. 1982. Shamans, Mystics and Doctors: A Psychological Inquiry into India and its Healing Traditions. New Delhi. Motilal Banarsidass
  • Kosambi, Damodar Dharmananda. 1962. Myth and Reality. Bombay. Popular Prakashan
  • Marriott, McKim ed,. 1990. Constructing an Indian ethnosociology. 1-39 India through Hindu Categories New Delhi. Sage Publications 10.1177/006996689023001003
  • Masilamani-Meyer, Eveline. 2004. Guardians of Tamilnadu: Folk Deities, Folk Religion, Hindu Themes. Halle. Verlag der Franckeschen Stiftungen zu Halle
  • Mullick, Samar Bosu. 2000. Gender relations and witches among the indigenous communities of Jharkhand, India. 4: 333-358 Gender, Technology and Development 10.1177/097185240000400301
  • Nabokov, Isabelle. 2000. Deadly power: A funeral to counter sorcery in south India. 27: 147-168 American Ethnologist 10.1525/ae.2000.27.1.147
  • Nandy, Ashis. 1989. Shamans, savages and the wilderness: On the audibility of dissent and the future of civilizations. 14: 147-168 Alternatives 10.1177/030437548901400301
  • Palriwala, Rajni, and Patricia Uberoi. 2008. Exploring the links: Gender issues in marriage and migration. 23-60 Marriage, Migration and Gender edited by Rajni Palriwala and Patricia Uberoi, New Delhi. Sage Publications 10.4135/9788132100324.n1
  • Quack, Johannes. 2012. Disenchanting India: Organized Rationalism and Criticism of Religion in India. Oxford. Oxford University Press 10.1093/acprof:oso/9780199812608.001.0001
  • Raheja, Gloria Goodwin, and Ann Grodzins Gold. 1994. Listen to the Heron's Words: Reimagining Gender and Kinship in North India. Berkeley. University of California Press
  • Rairkar, Hema. 2007. Midwives: A tradition on the move in Maharashtra. 413-472 Communication Processes, Volume 2: The Social and The Symbolic edited by Bernard Bel, Jan Brouwer, Biswajit Das, Vibodh Parthasarathi, and Guy Poetevin, New Delhi. Sage Publications 10.4135/9788132101178.n15
  • Sadgopal, Meera. 2009. Can maternity services open up to the indigenous traditions of midwifery?. 44: 52-59 Economic and Political Weekly
  • Sax, William S. 2010. Ritual and the problem of efficacy. 3-16 The Problem of Ritual Efficacy edited by William S. Sax, Johannes Quack, and Jan Weinhold, Oxford. Oxford University Press 10.1093/acprof:oso/9780195394405.003.0001
  • Scheper-Hughes, Nancy. 1992. Death Without Weeping: The Violence of Everyday Life in Brazil. Berkeley. University of California Press
  • Skaria, Ajay. 1997. Women, witchcraft and gratuitous violence in colonial western India. 155: 109-141 Past & Present 10.1093/past/155.1.109
  • Sundar, Nandini. 2001. Divining evil: The state and witchcraft in Bastar. 5: 425-448 Gender, Technology and Development 10.1177/097185240100500305
  • Vanashree. 2010. Witchcraft: Pain, resistance and the ceremony of punishment—Mahasweta Devi’s Bayen. 17: 223-247 Indian Journal of Gender Studies 10.1177/097152151001700202
  • Śarmā, Paṇḍit Kisanlāl. 2012. Āplā sampūrṇa cāturmās [Your Complete Four Holy Months], 4th ed.. Mumbai. Manoramā Prakāśan