(commented by Brick 2010, p. 206)

Only in the *Dharmanibandha* texts—a genre of later commentaries and digests, stemming from the twelfth to the nineteenth centuries—are there more extensive discussions on *sat¯ı*, both on its preconditions and on ritual instructions. As we saw above, the practice may have gained some ground between the ninth and eleventh centuries AD, among some locally confined martial classes. It may have gained a somewhat wider acclaim in the centuries leading up to the nineteenth century. To this we should add that gradually some Kashmiri and Bengali Brahmans appear to have adopted it in imitation of their rulers or for reasons of their own.

A tentative conclusion so far is that in pre-Vedic times there may or may not have been a custom of funerary gifts including the deceased's living wife or wives. In the Vedic passages referred to above, this may be hinted at as 'the ancient custom', although by then long extinct. But there may have been circles of ruling families, especially in situations of war, in which the custom of co-cremation was continued. Among aristocrats and in martial families there may have been individual cases, far and wide apart, of heroic self-sacrifice, such as by widowed queens whose honor, loyalty and courage at dramatic moments were admired but not generally emulated. The practice was never widespread, and never was it sanctioned unreservedly. At some point in time, probably towards the end of the first millennium CE, the topic began to enter debates about individual *dharma* (proper behavior for one's class, caste, and phase in life). In general, suicide was considered a grave sin, and so was homicide (forcing a woman to the pyre). At a relatively late stage there must have been a crucial turn among Brahmans, probably first in Kashmir and later in Bengal and Nepal. The reasons may have been varied: changes in property laws, changes in the minimum age of child brides, changes in the visions of the afterlife, possibly. Another factor may be that the position of high-class Hindu women had partially changed as a result of Islamic Mughal rule. In general, *sahagamana* as the noble example of a wife's loyalty began to be more favorably received, but as a practice it had never become rampant before European travelers began to report on it (Figueira 1994; Lewis 1994). Soon, it became expected from any foreign traveler to write an eyewitness account of such a case, but there is no clue to the exact number of cases before the beginning of the nineteenth century, when especially missionaries began to count the number of widow-burnings in their reports (Courtright 2011, pp. 170–71).

## **5. The Imperial–Colonial and Post-Independence Period**

The oldest material evidence of *sat¯ı* as an actual practice are inscriptions: one in Nepal dated 464 CE<sup>19</sup> and one in India dated 510 CE. There are surviving *sat¯ı* commemorative stones from the eleventh century onwards (Michaels 2004, p. 149). European travelers to South Asia, as far back as Marco Polo (late thirteenth century), made mention of the practice of widow burning, and later it became an almost obligatory item in the evolving genre of travel writing, whether as an eyewitness account or from hearsay. Whether these accounts testified to an actual flourishing of the practice, or tell us more about the European frisson about a cruel exotic culture, is not for us to decide. Whereas *sat¯ı* has always been the exception rather than the rule in Hindu life, it is striking that in the orientalizing imagination, religiously motivated suicide became regarded as one of the ubiquitous excesses both of the religious sentiment and of its authorities, the Brahmans. At the same time, some visitors could not help but express admiration for the noble women who did so voluntarily, although priests were often blamed for advocating the practice as a honorable 'way out' of dishonor, especially in those regions where widows had limited or no inheritance rights.20 In the Mughal period (1526–1857), the Muslim rulers had generally been averse to it: some (such as Akbar, in 1582) had merely prohibited the use of compulsion, others had actually banned the practice (see Figure 3).

In 1510, the famous Portuguese Alfonso Albequerque (governor of Portuguese India from 1509 to 1515) temporarily prohibited *sat¯ı* within the Portuguese territory of Goa, even though he had promised to respect local customs and to adopt a policy of noninterference in Hindu religious practices. This abolition was repealed in 1555 after Brahmanic intervention.21 Whether the practice had become more widespread autonomously, or whether the European and Mughal condemnation began to act as an incentive for 'going underground' and flourishing out of reach of imperial–colonial power, there is some numerical evidence that there were considerable amounts of cases in some districts before *sat¯ı* was officially banned in the year 1829,<sup>22</sup> by Lord Bentinck, with support from Indian reform intellectuals such as Ram Mohan Roy. Preceding this, the colonial court in Calcutta had gradually built up a well-documented case against *sat¯ı*. From 1805, it had ordered *pan. d. its* to investigate those textual authorities that were most often referred to as prescribing *sat¯ı* as a religious obligation and an ancient custom. The advisory opinions that were produced by the court Brahmans were nuanced: *sat¯ı* could not be classified as an absolute religious obligation, but there existed certain authoritative statements, such as on which widows qualified and on the merits bestowed on the widow who accompanied her deceased husband in the hereafter. Parallel to the juridical inquiries and the compilation of statistical data, some well-established reform-minded Brahman intellectuals discussed, in various pamphlets, the lack of religious legitimacy of the practice from the perspective of the Dharmashastric tradition. Both Roy, as a public intellectual, and a group of evangelical missionaries such as William Carey and James Pegg, denounced the practice of *sat¯ı* and opened a reformist attack on the position of Hindu women. The discourse gradually shifted towards the greater civilizing mission of the British Empire as an agent of reform, and although there was some discussion on the position of Hindu wives and widows, women themselves were conspicuously absent from those debates and had no voice in the matter. Once the abolition of *sat¯ı* had been achieved, opposition to the policy prohibiting it

<sup>19</sup> See note 27. For a recent picture of a *sat¯ı* gate in Nepal, see Figure 4. For an older photograph of that same gate, see (Michaels 2008, p. 139).

<sup>20</sup> This is not self-evident, however. It is correctly pointed out by some historians that the existence of inheritance rights for widows in some parts of India may have caused their in-laws to use more force on the widow to become a *sat¯ı*. This would ensure them that the family property would remain undivided. In this light, the suggestion inherent in the two Vedic texts quoted above achieves a new poignancy: if the widow was to remarry inside the family, preferably to her eldest brother-in-law, the family property would be likewise ensured. The custom of levirate may well have its basis in such material and/or legal conditions.

<sup>21</sup> This was parallelled by similar measures by the Dutch and French in their respective colonies in India. The Danes permitted it until the nineteenth century.

<sup>22</sup> Given the patchwork nature of British control over South Asian states, it was only in 1861 that a general ban was issued by Queen Victoria, whereas, for instance, it had been banned in the City of Calcutta already since 1803.

rather easily dissipated, but the entire process gave rise to a growing suspicion among the people that Hindu religious traditions were under attack. This state of mind fed into uneasiness about further 'Anglicization', such as the educational system, the missionaries, and the English language. It may be that in the independence movement, the nation had been imagined as a virtuous mother (Mother India) but it took solid postcolonial criticism as well as feminist and subaltern voices to more thoroughly deconstruct the '*sat¯ı* of the mind' (Spivak 1988). The same is valid for contemporary Hinduism in Nepal (see Figure 4).

India gained independence and full political sovereignty in 1947. In its penal code, the ban on *sat¯ı* was maintained, but in India custom, usage, and statutory law are so inextricably mixed up in personal law that it was only in 1987 that there was a general reformulation in the form of the Sati Prevention Act, in which 'the supporting, encouraging, forcing, glorifying or attempting to commit sati' was explicitly declared illegal and punishable. This legal step was rather suddenly induced and actuated by a new case of *sat¯ı* that took India by storm and was widely picked up by national as well as international media. In general *sat¯ı* was not only seen as a crime but as a relic of a dark and primitive past. There was a widespread notion that it had become completely extinct. In a dramatic way, this was contradicted by the voluntary death of a young woman named Roop Kanwar (18 years old) in Deorala, a town about 70 km north of Rajasthan's capital Jaipur.<sup>23</sup> She had been married for less than a year when her husband (24 years old) had suddenly died in hospital. It was told that immediately when she had heard about her husband's death, she had dressed up as a bride and had announced that she would follow him as a *sat¯ı*. <sup>24</sup> On the 4th of September 1987 she joined him on the pyre merely six hours after the corpse had been brought home from the hospital.25 It was only a matter of days before thousands of local people flocked to the place to receive a blessing. Naturally, all kinds of reporters, international journalists and researchers from women's organizations came flocking as well (Nandy 1994). There were massive counter-demonstrations for the right to widow-burning, mainly by women from the same state. The resulting court case languished for over twenty years: had her in-laws coerced her or adequately tried to dissuade her? Why had her own parents been absent? Had she been confused and overwhelmed? Had she made a mature decision all by herself? Was it her community, the Shekhawat Rajput caste in eastern Rajasthan, that had instilled such heroic romantic values in her? ¯ Was there an underground revival of the practice behind this?26

**Figure 4.** Pa´supatinath cremation ¯ *gha¯t.s*, Deopatan, Nepal, with 'Sat¯ıdvara' (the historical ¯ *sat¯ıs'* gate) indicated in red (photograph courtesy of Purushotam Khatiwada, Kathmandu).

In several ways, women's organizations took the lead in interpreting this shocking recurrence of *sat¯ı*: the case was presented as shaming India's modernity and as an illustration of the tenacious hold that patriarchal institutions and caste values still had on young naïve minds—and, as was said explicitly, on female bodies (Oldenburg 1994b). Although some forms of Indian feminism were explicitly against

<sup>23</sup> Whereas there have been rumors about some other cases since independence, this is certainly the most high-profile *sat¯ı* incident.

<sup>24</sup> For some locally situated research, see (Van Den Bosch 1990, 1992, 1995). For focus on the widow, see (Buitelaar 1995).

<sup>25</sup> In this contemporary case it is fortunate that we have this detail. Without claiming that this time lapse of a mere six hours is representative for most cases, it illustrates the extremely short time the widow may have had for her deliberation, and for the parents or parents-in-law to support or dissuade her. Moreover, it remains completely invisible how despite political and judicial processes in the outside world, certain tendencies and contests might have formed within households and within individuals.

<sup>26</sup> Even as recently as the 5th of September, 2019, on the 32nd anniversary of the event, *The Times of India* reported on the 'court case' with the headline 'In Rajasthan Roop Kanwar still burns bright', and placed a photograph of her *samadhi sthal ¯* (the monument marking the place where she was burnt).

modeling themselves after 'Western' feminism, not only the practice but also the very ontology behind *sat¯ı* were generally seen as blatantly harmful and opposed to women's well-being. As in the ages before, this singular case evoked contradictory trajectories of interpretation and evaluation: a horrid rite symbolizing abuse of women by relations, in-laws, priests, and caste communities, but just as well the romantic appeal of true love and conjugal loyalty. Others used the occasion to unfold critical theories about female agency. Particularly, Gayatri Spivak became a spokesperson of/for what she called 'sub-altern voices', asking herself and others: 'Can the subaltern speak?' (Spivak 1988). In this case, the subaltern is the disenfranchised woman whose subaltern voice had remained mostly unheard and whose arguments had remained marginal to the debate.<sup>27</sup> Closer historical investigation revealed, however, that women in the past, caught between bleak widowhood and the flames of the pyre, had had a voice, albeit a muted one, instrumentalized by evangelical missionaries for their own ends. It was found that amidst the multivocality of millennia of opiniated male players in the grim drama of burning widows, colonial records of the years between 1805 and 1830 testified that those women who had managed to escape coercion or had been rescued by British officers, respectively talked around by Indian reformists, gave as their prime motive for co-cremation: material suffering (Doniger 2010, p. 613). As subjective dignity of the women involved—and particularly those who had died in the flames—this may not amount to much, but it strengthened the urgency with which critical theories of the Frankfurter Schule were applied to both women's issues and postcolonial issues in the long shadows of 'inner colonization'.
