**1. Introduction**

In the eighteenth and nineteenth centuries, the spectacular phenomenon of Indian women committing ritual suicide by entering their dead husbands' funeral pyres strained the limits of cross-cultural understanding. When the British conquered India, they came face to face with practices they could neither understand nor condone, and in the Indo-British colonial encounter, this occasional funerary practice became one of the prime examples of what they considered the 'horrid rites of Hindustan'. Yet, there was European fascination as well, mostly with the bravery of the woman. We have found this ambivalence expressed through early travelers' accounts, ranging from ancient 'historiographies' (Karttunen 1989; 1997, pp. 64–67); through Venetian traveler–merchant Marco Polo's *Il Milione* (put on paper by Rustichello in 1298) and Portuguese General Albuquerque (the later Viceroy), who managed to temporarily abolish the practice around Goa in 1510 (Major 2006, 2011); and to British merchant missionaries, whose role in abolition was merely an offshoot of a much grander design, the conversion to Enlightenment rationality and/or Christianity (Mani 1998). Whereas, from the European perspective, widow-burning became an alibi for the colonial civilizing mission, by the turn of the nineteenth to the twentieth century, it also became a potent signifier for the oppression of women and the degradation of the once-great Indian civilization as a whole. The colonial divide necessitated a complex mediation, structured not only by relations of domination and subordination, but also by anxieties about the so-called noninterference. It was a fundamental principle of the British government to exercise 'toleration' in matters of religion, and to the orientalizing imagination, India was a place where religion held unbridled sway over both feelings and facts (Figure 1). In this context, the rhetoric

of reform had a wobbly basis, to say the least. Although the desirability of legislative prohibition was increasingly supported by indigenous reformers as well, especially in the state of Bengal (Ramos 2017), co-cremation was otherwise defended as an inviolable aspect of Hindu tradition and as a religiously sanctioned ritual (Weinberger-Thomas 1999). While during British colonial rule in India the widow tended to be portrayed as a tragic victim, her own voice and agency were shockingly marginalized and practically absent in all descriptions, deliberations, and decisions.

**Figure 1.** 'Widow Burning in India'; illustration by the Wesleyan Missionary Society (a public domain image, https://archive.org/downloads/wesleyanjuvenil08socigoog/wesleyanjuvenil08socigoog.pdf).

In the above lines, the tendency to regard ritual acts as a model of and a model for a particular cultural world—that of Hindu subjects under colonial rule—is obvious. To counterbalance this, we should at least make an attempt to understand the practice as a ritual in its own right as well (Handelman 2004; Kapferer 2004). In addition to a representation of co-cremation as some trait essential to 'eternal' Hinduism (*sanatana dharma ¯* ), or as an arena in which colonial and anticolonial forces were played out along with the related conflicts of social, economic, and political rivalry, the ritual practice should be afforded its own integrity as well. Accordingly, we should allow the interior dynamics of this ritual to speak not only of the underlying sociocultural order, but also of its own autonomous position distinct from such an order; we should prevent such a presentation from slipping into an inherently and exclusively functionalist interpretation. At the end of this article we will have shown that, as rituals, both *sat¯ı* and accepting widowhood required rites of passage in which the ritual actions were more than representations: Hindu widows most literally had to cross a threshold in a matter of hours. Without that step, the widowed wife would remain in a liminal situation for the rest of her life.

Under various names, there may have been, in India, an ancient but never widely practiced ritual technically (and possibly euphemistically) related to as co-cremation,1 but culture-specifically better known as *sahamaran. a* (dying together), *sahagaman. a* (going together), *anumaran. a* (following in death),

<sup>1</sup> Or con-cremation, the term used by M. Monier-Williams, in *Indian Wisdom* (Williams 1875, p. 258, fn. 2).

*anugaman. a* (going along with or after someone), *anumr.tyu* (the practice of following someone in death), and *anvarohana ¯* (climbing the pyre after someone). It was only much later (in 1787) that the word *sat¯ı* (indicating both the practice and the woman who had performed it) came into vogue; the British typically made this into the infamous 'suttee'. A few other fringe names exist, such as *me-satya* in some Hindu milieus of colonial Bali (rumored to be the proud exclamation of the widow: 'mine is the truth'). The practice of co-cremation, as an anthropologically neutral and overarching term, may be compared to other related practices with similar technical terms, comprising various time-, space- and ethnically specific practices of co-death or co-disposal, such as co-interment (being buried together) and co-immersion (being immersed together in a body of (mostly flowing) water). Those generic technical terms do not necessarily indicate conjugality, like they do in the Hindu case, nor do they indicate the imbalance that one had died first and that the other person followed him alive, either willingly or unwillingly. War and epidemics all over the world may long have been occasions for any type of joint or even mass disposals, and romance anywhere may have induced an occasional self-chosen death after a partner's demise. In Hindu India,2 there has indeed been a tendency that such practices gained a certain prestige in ruling families, in which both the code of honor and the code of property may have inspired the customary right (respectively duty) to accompany the deceased male to the beyond.

How this Hindu conjugal practice—a widow following her deceased husband to the funeral pyre and being cremated alive with him—came to have some acclaim is not supported by a singular unilinear story. Instead, we have a fragmented narrative through the ages, with an overrepresentation of 'foreign' reports: alleged eyewitness accounts, written by Greek historiographers, early European and Middle Eastern merchants, Mughal rulers (1526–1857), and imperial–colonial travelers. As a first introductory window into the practice, we draw attention to the widely spread (both in time and space) etymological connotation of the word 'widow'. The contemporary English word has an all-over Indo-European lineage, and goes back to a Proto-Indo-European verb (\*uidh), meaning to (be) separate, to (be) empty, to (be) divide(d), as well as to be destitute, to lack.3 Even now, the Sanskrit word for widow (*vidhava¯*) signifies lonely, solitary, bereft, and more specifically: husbandless. Although the same is valid for the widower, it has long been customary law among Hindus that widowers could easily remarry, whereas widows at times and in certain milieus could not. This imbalance in perspectives may have some explanatory power from a socioeconomic point of view. The salience of widow-burning over time has indeed much depended on the conditions of whether widows in particular regions and under particular regimes had the right to inherit and/or remarry, or not.4

At the end of the first millennium CE, possibly as a response to a growing critique (both internally and externally), some defenders of the practice of self-immolation of widows, schooled in Vedic and Hindu scriptures, began to point at authoritative texts as well. In defense of co-cremation, a cryptic and highly ambiguous text from the *Atharvaveda* (AV 18.3.1)5 was often referred to as the locus classicus:

Choosing the husband's world, this woman lays herself down beside his lifeless body. Faithfully preserving the ancient custom, bestow upon her both wealth and offspring.

But if one continues reading (AV 18.3.2-8), one can't help but notice the dramatic performative reversal:

<sup>2</sup> In Hindu philosophy, but also as a basis for Hindu common sense in general, there is the acknowledgement that every phenomenon should be understood in the light of *de´sa-kala-p ¯ atra ¯* , meaning: in the light of place (country, habitat, ambiance, atmosphere, occasion), time (proper time, time period, age, phase, cosmic cycle), and the subject (person, performer, the individual in his/her circumstances). Applying this to the phenomenon of *sat¯ı* would result in a welcome de-essentializing of the practice, recognizing also that it has been performed relatively rarely and has always been contested. For the line of argumentation in this article, it is crucial to emphasize that the practice has been banned for almost 200 years. See also notes 7, 31, and 42.

<sup>3</sup> Compare Latin *vidua*, German *Witwe*, and French *veuve*.

<sup>4</sup> See note 18, below.

<sup>5</sup> With a parallel in *R. gveda* 10.18.7-8; see further below.

Come, rise unto the world of the living, o woman, since the man by whose side you now lie, is lifeless. [ ... ] Rise, abandon the dead, re-join the living.

Evidently, there is more to this than meets the eye: was it meant to be a merely symbolic gesture—a performance, a ritual re-enactment possibly based on a long extinct archaic practice—when she joined her deceased husband for a moment? Was she meant to return to the world of the living, the social world? If so, what is hinted at with the contradictory promise of wealth and offspring in the first stanza? How could the gods (and their spokespersons, the funeral priests) promise her wealth and offspring when she was to be cut off from life? Those and other textual enigmas will be explored in Section 3. 6

In Section 2 we describe widows' co-cremation according to its ritual aspects. In this, we try to refrain from interpretation and contextualization in an attempt to present co-cremation in the bare ritual components traditionally known to us.

In Section 3 we contextualize the phenomenon: the ever-returning trope of sacrifice; as well as matters of widowhood, iconography, auspiciousness, and a woman's *dharma* as a wife and a widow.

In Section 4 we explore some textual passages covering more than two millennia.

In Section 5 we zoom in on British and independent India: *sat¯ı*/suttee as a political issue; the case of Roop Kanwar in 1987, and the now extremely rare ritual of *sat¯ı* as still constituting a contradictory highly ambiguous interpretive category.

In Sections 6–8, some tentative conclusions and critical considerations are given; also, we make a plea for a place- and time-sensitive type of ritual criticism.

This structure required a multimethodological approach. In order to let a ritual—and its ritual variations—speak for itself first, we composed a collage of ritual components, as they may have determined the procedure itself, with some of its variations through time and space. Apart from ritual studies and religious studies, we used insights from performance studies as well as from the study of material culture, such as relating to the shrines, votive objects, and goddess worship. For the phenomenon's contextualization we took recourse to the type of texts that are most often quoted as either inspiring the practice or contesting it. Apart from the religious 'validity' of the practice as allegedly found in ancient texts, and later commentaries or digests, we needed to look into the social and legal realities of women within those milieus in which the practice of a widow's self-immolation has had some sanction. In order to understand how the 'foreign male gaze', from Alexander the Great's 'historiographers' and 'ethnographers' to contemporary feminists, journalists, and comparative ethicists, has impacted the formulation of both its inspiring and its horrifying character, we needed insights from travel writing, orientalism, feminism, and subaltern studies.

Yet, it goes almost without saying that all such forays are subservient to the central question that we hoped to contribute to this Special Issue on contemporary ritual: was it justified, and cross-culturally fair, when Ronald Grimes, with his vast readership among scholars of rituals, discussed India's long extinct practice of widows' co-cremation as a topic of contemporary ritual criticism?

<sup>6</sup> Brick (2010, p. 205): 'Considering the complete absence of any mention of *sahagamana* in both Vedic literature and the earliest works of the *Dharma´sastra ¯* tradition, it seems reasonable to conclude that this practice first gained enough popularity within Brahmanical culture to warrant mention at approximately the time when [ ... ] composed their works on *dharma'*, broadly in the second half of the first millennium CE. For historical accuracy, it is crucial to note here that by far not all Sanskrit authors who mentioned the practice expressed their opinion, let alone endorsed it as a meritorious alternative to ascetic celibacy for at least some women. If a position was taken in these earlier commentaries and digests, the practice was often refuted. At some point much later, commentators in late Dharmashastric texts of the medieval period, who were obviously directly engaged with one another in a complex pan-Indian discourse, began to refer the practice to Vedic scriptures. Locus classicus was the cryptic and variously translated Ayurvedic passage quoted above and its Rigvedic parallel cq precursor in RV 10.18.7. Such reference to alleged Vedic sanction was often made only indirectly, via a problematic passage in the *Brahmapura¯n. a*, probably in an attempt to find Vedic support for the practice. The name of the commentator most strongly linked to this 'Vedic argument' is Raghunandana (c. 1500 AD), who may have either misread the text or purposely falsified RV 10.18.7. See also Section 4.

Although the platform of ritual studies is known for its plurality of methods and its multidisciplinarity, every author has a pedigree and a distinctive mark of one's own. In this article we took the position of the in-between researcher, who had been thoroughly trained in South Asian languages and cultures, and gradually began to move over to disciplinary fields such as religious studies and ritual studies. To be honest, the initial trigger had been confusion, and then irritation: why 'freeze' and 'essentialize' a long-extinct distant practice, and propose it as a topic for ritual criticism on a par with current practices of clitoridectomy? For this research we used our South-Asian expertise to sketch the settings in which Hindu practices of female co-cremation historically occurred, and we use insights into ritual performances to zoom in on *sat¯ı* as one of three ways traditionally open to a Hindu widow. Due to the fact that we have to 'set something straight' at this cross-cultural frontier after three decades of ritual criticism, during which the topic of co-cremation rather lamely and lopsidedly hobbled along,<sup>7</sup> we end this article with a few suggestions for a more embodied and more embedded perspective.
