Defying blessings of the goddess and the community: Disputes over sati (widow burning) in contemporary India
Masakazu Tanaka.

1.Introduction.
This article aims to make clear the nature of ritual violence through the analysis of disputes over sati in contemporary India, pointing out the difficulties in defying it and investigating the possibilities of overcoming it. Originally, Sati referred to chaste women and the goddess symbolising chastity and not to the custom of Hindu widows being burned or buried with the husbandfs corpse(1). It was the British colonial rulers who understood the word sati as the act of burning widows. The idea is that the life on this earth literally comes to an end there and the husband and wife live together for ever in heaven. The British prohibited sati as a brutal custom in the beginning of the nineteenth century. Many researchers have already attempted analyses of sati in those days and of the disputes leading to the prohibition of sati(2). Here I deal with the sati that took place in India in 1987 and its repercussions. I had visited the village concerned around six months after the incident, but the atmosphere was still such that no research could be conducted openly(3). For this reason, newspaper and magazine articles on the incident will be used here as the objects of analysis. The number of articles on this topic comes to about two hundred ranging from those that just report the facts of the sati incident and the names of people arrested to special reports of several pages in length(4). In addition to these there are many research articles by historians, religious studies specialists and anthropologists that have been published till recently. In this article, I first present a brief account of the incident in 1987 and then analyse the disputes over sati appearing in newspapers. What I want to focus on in particular are the beliefs behind sati, the pro-sati views of those who praise and defend sati and the anti-sati views of those who criticise it as an evil tradition of the past.
Ritual violence refers to two things: patterned or formalised violence and violent elements in ritual. In other words this is community violence in the sense that it is socially sanctioned(5), and it is divine violence in the sense that it is linked to a supernatural being. As we will see later, sati is also ritualised violence which is worshipped by many people and blessed by the goddess. Widows who can receive the blessings of the goddess are limited. The widow must also show a miracle in the act of sati. Once this is accepted those who oppose the sati are cursed by the divine beings. Thus to reproach sati means the same as criticising the community and its system of belief. This is not just a matter of criticising violence as evil, since in order to criticise it, the community and the supernatural beings have to be reprimanded. Disputes over sati teach us the difficulties in this kind of criticisms of violence.

2. Deorala, a Site of sati
On 4th September 1987 in Rajasthan, a state in northeast India, a young woman was cremated along with her husband who had died a few hours ago. To be more accurate, the word cremated applies only to the husband since the woman was still alive when she was burned. The first article in an English language newspaper reporting this incident was only a short one of about a hundred words (Indian Express 6th September 1987)(6). In an article "A young woman commits sati" dated 5th September, it is said the police reported that an eighteen-year-old woman has committed sati and that four relatives have been arrested for aiding the suicide. There is no comment in particular.
Let me first present what were reported as facts. The village of Deorala is located eighty kilometres north of the state capital Jaipur and just twenty kilometres off the main road leading to Delhi, the capital city and it is by no means an isolated village in the middle of nowhere. The population is 13,000 and there are 300 households of Rajputs who are the dominant caste and own 70 percent of the land in the entire village. The village is well off, it has five primary schools and the level of education is higher than in other regions(7).
Roop Kanwar's sati was the fourth one in this village as far as it is known and the last sati took place seventy years ago. Thirty-eight people have committed sati in Rajasthan since Independence and all have been in the Shekawati district which Deorala is a part of.
Roop Kanwar was eighteen years old, when she married the twenty-four year old Man Singh on 17th January 1987. The husband did not have a job. He had been trying to become a medical student, and had taken a university entrance examination before and after the marriage. He was taken into hospital on 3rd September with a stomach pain, but he died next morning(8). This happened eight months after the marriage. Taking into account the period that she had spent in her fathers house after marriage, Roop Kanwar and Man Singhs time together seems to have been much shorter - some say just three weeks. She saw her husband's corpse and expressed her will to perform sati. She immediately put on her red wedding clothes and went with the funeral procession to the cremation ground. This was around one o'clock in the afternoon. As the villagers watched, she sat on the wood piled up to burn the corpse. The fifteen year-old younger brother of Singh lit the funeral pyre as the chief mourner(9). There were said to have been four to five thousand spectators(10). Roop Kanwars parental house got to know about their daughters sati in the newspapers next morning.
Roop Kanwar was born in a well-to-do family in Jaipur, the capital city of Rajasthan. Her father ran a transportation business there. She was born and brought up in Ranchi in the state of Bihar, and there was a temple of sati nearby. Speculations that this might have influenced her were also reported. She was educated up to eighth class in Ranchi. She went on to high school in Jaipur, but she gave up her studies for marriage. Her father-in-law was an English teacher in Deorala village.
Roop Kanwar's sati attracted neighbouring villagers and several thousand worshippers visited every day even after the incident. The cremation place was called sati stal and was covered by a yellow canopy mixed with light pink. Four Rajput youths with open swords stood in front of the place and twenty-seven youths circled it without stop, guarding the cremation ashes twenty-four hours a day(11).
The number of participants reached a climax on the morning of 16th September, when the ritual ending the mourning period called the chunari festival took place. A chunari (an expensive gold-embroidered silk shawl worn by a bride at her wedding) was taken along with other items from Roop Kanwar's natal house to her in-laws house and then to the cremation ground. There the chunari was burned in the sacred fire amidst the crowd shouting "Victory to goddess Sati!" and priests chanting the Gayatri mantra. The priests cooled the ashes down, gathered them in a pot and buried it. It was said that the ashes were saturated with sacred power (sakti). A trident was positioned where it was buried, worship was performed and more chunari shawls were placed on top. Around 300,000 people participated and some had even walked barefoot from six hundred kilometres away
. Many souvenir shops were set up and many composite pictures were sold of Roop Kanwar and her husband showing the couple enveloped in flames. Even after the chunari festival, two to three thousand people came to Deorala every day. In Roop Kanwars house the wedding photograph was displayed on the veranda and in the room where she decided to commit sati, the red and silver sari which she was wearing at the time was spread out on display. Elite members of the village decided to collect funds to build a temple for Roop Kanwar at the place of cremation. It is said 2,000,000 - 3,000,000 yen was collected just on the day of the chunari festival.
Women activists reminded the Chief Minister of Rajasthan of the anti-sati act of 1827, demanding an explanation from him, but were turned away. On 14th September, around three hundred and fifty members of thirteen womens organisations took part in a silent demonstration in Jaipur criticising the government for not taking any action on 14th September. They took the matter to the Rajasthan High Court and on 15th September an order was passed banning the chunari festival. However, this ban was not implemented, the government remained silent and there were even some members of the state assembly who participated in it. At this point, only Roop Kanwars younger brother-in-law who had lit the funeral pyre had been arrested. On 21st September a demonstration was organised in Delhi demanding that the Chief Minister should resign and that the central government should prevent the construction of the temple. On 26th over fifty widows joined a demonstration against sati and government relief measures for widows.
Rajiv Gandhi who was the Prime Minister at the time sent a minister on 22nd September for investigation into the incident and statement of the central governments position. The state government reacted under pressure, and on 19th September the father-in-law and three other relatives were arrested for aiding suicide. Twenty-eight people were arrested by 27th September. Policemen were posted in the village and all souvenir shops were banned by 2nd October. The state government issued a law prohibiting the practice and encouragement of sati on 1st October. On 20th October the police demolished a Roop Kanwar monument and on 31st October they arrested five people belonging to the Association for Protection of the Righteousness of Sati (Sati Dharma Rakshya Samiti) for encouraging the glorification of sati.
Against these moves of the anti-sati lobby, the pro-sati groups responded by organising a demonstration of around five hundred Rajput men in Jaipur on 19th September as a protest against the arrests made on that day. They demanded that the government should not interfere in religious matters. On 27th September the first meeting of the Association for Protection of the Righteousness of Sati was held in Deorala, and the core members were the Rajputs living in Jaipur. On 8th October they hosted a pro-sati demonstration in Jaipur of about fifty thousand people, which was the largest demonstration organised after the incident. Kalyan Singh Kalvi who was a central figure in the Association for Protection of the Righteousness of Sati and the leader of the states Janata Party also participated and gave a speech at the demonstration.
The influence of anti-sati movements went beyond Rajasthan and on 2nd October a womens organisation criticised the procession at a festival of a temple dedicated to sati in Bombay. Festivals were also banned in a sati temple in Calcutta. Thirty-six womens organisations demonstrated in Delhi on 4th November demanding the establishment of a law prohibiting sati for the whole of India. In response to this, a new prohibition law was passed in the Lower House and Upper House by 16th December. The Chief Minister of Rajasthan resigned on 18th January, and one of the reasons is said to be his inadequate response to Roop Kanwars sati incident. However, those arrested were later released without bail and no trials have been carried out. Furthermore, in spite of the legal regulations, eight thousand people gathered in Deorala for the first anniversary of Roop Kanwars death.
Let me make a brief summary of the points of dispute. First of all, the pro-sati viewpoint sees sati as an honoured tradition particular to the Rajputs or Hindus and is antagonistic to the state. By considering sati as religious, it criticises state intervention. The anti-sati viewpoint held most strongly by womens organisations, criticises sati as savage and emphasises the importance of education and the improvement in status of widows. For them, sati is a womens problem and a custom that symbolises the low status of women in India. There is no question of it being a miraculous divine deed and there is no such thing as voluntary sati. They argue that sati cannot be said to be religious and that it is a means of making money at the expense of women. Their attack is also directed against the government as they declare that the government is not sincerely trying to outlaw sati and that the legal prohibitions are merely a gesture. But what they actually did was to call for government action. They not only demanded the urgent arrest of the relatives involved in Roop Kanwar's sati, but also acted against temples commemorating sati in the past in Calcutta and Bombay, saying that they glorify sati.

3. Sati as miracle
Here I would like to consider the beliefs behind sati. The anthropologist, Harlan gives the following explanation(12). In Rajasthan, the value of a woman is determined by her chastity towards the husband (pati). By this the husband is protected from all kinds of danger. The practice of this value is called pativrata. The woman performs another practice when the husband dies. This is sativrata and by this her body gets hot. That is why it is said that when she chooses her death on the funeral pyre, she does not feel the heat of the flames at all. When she dies, she is called sati mata (Sati goddess). Sati is the manifestation of the moral goodness (sat) in women. In the newspapers, sat is expressed as something that takes hold of a widow, or it is explained that only a widow who has received a revelation from the goddess leaves everything behind and performs sati (Telegraph 10th October 1987).
Let us take a more detailed look at the process between the husbands death and the act of sati. When the widow tells her family of her intension to commit sati, she puts her finger near a candle flame and proves that it does not get burned. Next she takes a bath in a river and changes into her bridal clothes. If it is raining, it will stop and the suns rays will shine on the funeral pyre. The widow jumps on to the pile of wood. It is said that the husband comes to life for a moment to acknowledge his wifes sati. No one lights the pyre. Flames appear from the heat of the widows body itself, as she receives the goddess's blessing. For the widow, it feels like bathing in water and is literally called "bathing in flames" (agni snana).
Roop Kanwar was filled with power and due to this power she was able to perform sati. Then, she became a goddess and gained the power to grant the wishes of other people. People came into contact with this sacred power at the cremation ground and sought for all kinds of worldly benefits.
From the point of view of other women, Roop Kanwars sati signified her exceptional strength. The power of the widow who has decided to perform sati appears as a blessing and at the same time a curse to people. People cannot stop sati in fear of her curse. It is said that the lineage of the man who tried to prevent the sati in Deorala seventy years ago came to an end after that. It is also said that the policeman who arrested Roop Kanwars father-in-law fell ill.
Whatever the status given to sati in the scriptures, the explanations above seem to be enough to tell us about the beliefs that bring about sati. However, there is a contradiction here. For instance, a woman who commits sati is destined to be so, that means she is a chaste woman who has always been serving her husband. But, if that is the case, why does the husband die before her cannot be explained? Why did she not die before her husband? This is surely a major contradiction if we take into consideration the fact that a widow is often said to be the cause of her husbands death. Moreover, emphasis on the miracle may also lead to the denial of the sati that actually occurred. As some of the villagers point out, the fact that it was the younger brother-in-law who lit the funeral pyre and that there were some people arrested in Roop Kanwars sati led to the denial of its authenticity (Times of India 11th December 1987). There is also a view that sat could not have manifested in Roop Kanwar since she was too young(13).
In this way, the logic of the people supporting sati seems comprehensive at first sight if we accept the concepts that are presupposed, such as the existence of sat and its power. But it is by no means so, and even if the beliefs themselves were not denied by facts, the authenticity of each of the incidents would be questioned. Roop Kanwars sati was also not exempt from this inspection.
Furthermore, it is believed that the one who commits sati takes away not only the sins of her own, but also those of her husband, the ancestors and the descendents. This is a belief that deeply reflects the sacrificial nature of sati. I heard for instance that in the case of Roop Kanwars sati, there was plenty of rain and abundant crops that year thanks to her. It would not be impossible here to identify the aspect of human sacrifice. There are various aspects to sati that cannot be understood simply as beliefs regarding chaste women.

4. Sati as resistance
Roop Kanwars father said that he was proud of what his daughter had done. She has made both our family and the in-laws family immortal (Indian Express 11th September 1987). But this was not all. As we will see below, her death can be interpreted as not only making her family and relatives immortal, but also as being used politically to make the Rajputs and their land Rajasthan immortal.
As we have already seen, from the beginning of the sati incident, there was a conspicuous attempt to understand and defend this in relation to the Rajput caste. Demonstrations were organised many times with the Rajputs as the core. Why the Rajputs? If we take note of the fact that sati is a phenomenon particular to the Rajputs of Rajasthan, it is necessary to consider the conditions that the Rajputs are in when thinking about contemporary sati. Rajputs are a royal caste of northwestern India and were highly regarded as warriors even in the colonial period. They used to rule over Rajasthan, but their power has been weakened in recent years, externally due to the emergence of other castes and internally due to land reforms(14). This is a crisis for the Rajputs. Sati originates from the act of women committing suicide after losing their husbands in battle to prevent being shamed by the enemy, thus, is a suitable symbol for demonstrating their past glory both internally and externally. Sati can be seen as a means of resistance to overcome the crisis of Rajput identity or tradition(15).
A member of a pro-sati organisation, Kalyan Singh Kalvi, did not stop defending sati as a part of his (Rajput) tradition, in spite of the criticisms of sati by the Congress party members in the centre (Times of India 16th September 1987). Kalvi later weakened his view a little and said that the Rajasthan state government should have deterred sati in accordance with the law, but it was not right that the government should intervene in the chunari festival, and the construction of a memorial since these were only a part of family rituals (Times of India 29th September 1987). Here the relationship between politics and religion and that between the state and the family are put to question. Moreover, he even went on to state that the actions of the government and the police were wounding the religious sentiments of the Hindus, especially the Rajputs, and to hurt their pride would be very dangerous for the country since the Rajputs are emotional people and it is because of this that they have produced many martyrs as soldiers of the country (Times of India 30th September 1987).
Other pro-sati people emphasise similar points. According to them, enforced sati is a crime and should be reprimanded, but voluntary sati cannot be opposed or deterred. To do so, they say, would be government intervention in religious matters. They state that matters regarding religion should not be decided by the government or the courts, but by religious men such as the Shankararcharyas who head the four most important Hindu monasteries in India.
In this way, the defence of sati is reconsidered as the tradition of not only Rajputs but of Hindus in general, and its authority is sought in a sphere outside secular power. Here lies the reason why many articles point to the relationship between defenders of sati and contemporary Hindu nationalism, which calls to make Hinduism the state religion and India a Hindu state. It is thought that just as religion is only a tool for gaining political power in Hindu nationalism, sati is also only a tool in the power politics for politicians(16).
What I want to emphasise here is the point that it is not only because sati is a custom particular to India that it becomes a symbol. Not only this, but also more generally there is a historical relationship between Indian women and Hindu nationalism in the past. That is to say, in India as the anti-British movements became popular at the end of the nineteenth century, it came to be understood that women were the source and storehouse of Indian tradition. Unlike the men who worked outside, women who stayed at home were transformed into unpolluted beings, who were not influenced by the effects of modernisation and westernisation and were the basis of nationalism dreaming for a future independent state. Here the women who were idealised belonged to the upper castes observing social segregation (purdah). It was not only the British and Indian men who were put in opposition to this category. Westernised Hindu women and lower caste women who had no choice but to work outside along with men were also compared to these upper caste women. In this way, although it was paradoxical, a discourse was born according to which subjectivity was granted to the very women whose place was inside(17).
A similar rhetoric can be seen the dispute regarding Roop Kanwar's sati. That is to say, sati contains the paradox that it recognises true subjectivity in women who are subjugated to their husbands. Moreover, sati differentiates Rajput women from other low caste women in the region ("Their sati is not real.")(18) and at the same time it functions to negate the urban and westernised women and the secularism that they represent. Sati becomes the ultimate tradition to be observed by Rajput women. This tradition is ultimate for women in the sense that they are its true heirs and thus to prove themselves as Rajput women they have to choose death. Their subjectivity can only be actualised by death.

5. Resistance against sati
Most of the English language newspapers reporting Roop Kanwar's sati criticises it(19). For instance, Indian Express dated 11th September published an editorial with a title saying that a "barbarian tradition" has come back to life. In the editorial titled "Savage and primitive" published on the fourteenth day after the incident when the chunari ritual came to and end, it pointed out that, despite being an educated man, the father-in-law did not prevent the sati nor think that it was against the law. It also states that Roop Kanwar's death is related to violence against women, such as dowry murders and abortion of female foetuses after testing for its sex. It deplores that this incident proves India to be still very far from the twenty-first century. In other newspapers, sati is seen as "savage act" (Hindu 17th September 1987) and expressions such as "shame" and "inhuman act" are used and the word superstition is repeatedly employed (Hindustan Times 18th September 1987). The many opinions published in the newspapers were much the same. What is common in these editorials is the significance of education and administrative leadership beyond political interests. They criticise that the state government did hardly anything after the sati occurred in fear of what might happen to the results of the next elections. They also approve of the establishment of the new legal measures, but question the efficacy of the law.
Sati was understood above all as the symbol of the pathetic condition of widows in India. The fact that demonstrations by widows were organised points to this. In Hindu society, particularly among upper castes, widows are considered to be most inauspicious(20). This is because the cause of the husbands death is thought to be the wife who is still alive. Widow remarriage is accepted only among low castes. When a woman becomes a widow, various measures are taken to hide her sexuality which is potentially dangerous(21). The widow must shave the long hair which is a symbol of womanhood, take off all her jewellery and wear white or plain light colour sari. She is prohibited from taking her meals with other family members and is not allowed to participate in her own sons wedding. She is considered very inauspicious and so is not fit for an auspicious occasion such as marriage. A married woman who dies before her husband is the complete opposite to the widow. If the couple has a son then she is perfect. A married woman whose husband is alive has important roles in rituals, such as marriage, as an auspicious being. So it was said that if there were no such discrimination against widows, no one would want to commit sati.
Moreover, sati is said to be a means of deterring the wifes inheritance right. Especially in the case of Roop Kanwar who had no children, the large amount of dowry in cash and kind given to the husband at the time of the wedding was supposed to be returned when she became a widow(22). For this reason it was also interpreted and reported that the in-laws who did not want to do so forced the sati to take place.
There was also the criticism that contemporary sati is not a honourable tradition, but merely a religious show put up by the entire village aiming to gain the enormous amount of donations that would come from it. It was argued that this is not tradition but a contemporary phenomenon in a society penetrated by the market economy(23).
Furthermore, there were doubts regarding the incident itself. In a newspaper dated 25th September, a teacher who had taught Roop Kanwar in Ranchi was reported to have said that she was forced to sit on the funeral pyre and that she did not seem particularly religious when she was a student (Times of India 25th September 1987). But, the fact that she was forced has not been proved. A newspaper dated 29th September expresses doubts whether Roop Kanwar really committed sati of her own accord (Hindu 29th September 1987). According to one view, when Man Singh's body was carried into the house, the members of the household blamed Roop Kanwar that she had brought misfortune to the house and said that even greater misfortunes will befall the house in the future. It was written that even if she had tried to escape, it would have been impossible due to the way in which the wood was piled up and her screaming would not have been heard due to the cheering of the crowds. Later, on 2nd October, a police report was published saying that Roop Kanwar had been trying to escape from the flames that were burning her body.
Reports based on actual statements began to appear only after mid October. First, a report was published by three representatives of the Bombay Union of Journalists (Indian Express 23rd October 1987)(24). There it was pointed out that Roop Kanwar was hiding in a shed. After this, the tendency to discuss Roop Kanwars sati as murder grew stronger(25). However, even in these reports the truth is obscure.
The above criticisms moved the government, but were nullified in the face of enthusiasm of those actually involved. Those who were against sati were to face some dilemmas themselves. These were the problems concerning the subjectivity of women and the fact that, however justified their argument was, it nonetheless could lead to supporting the control and punishment of minority groups, the Rajputs here, by state power.
To totally deny the existence of voluntary sati would be to accept the view of women as being powerless and as victims. In the case of sati the object of violence is a woman. But to criticise sati as the ultimate violence of patriarchal society and to talk of women as victims of evil customs could lead to denying the nature of women as agents from the start. As a result, one ends up supporting the typical view of women in patriarchal society, that they are dependent and cannot decide anything for themselves. It would then seem that the pro-sati discourse praising the heroism of women who commit sati was recognising the subjectivity of women, though it may not apply to women in general, as it denies the agency of those widows who have failed to commit sati.
On the other hand, if they accept voluntary sati, not only would they be unable to deny sati itself, but they would also be strengthening the image of women as being tied to tradition in this case as well. This is because the decision to commit sati is made by a woman who wants to be subjugated to her husband and sacrifice everything for him. This clearly shows that sati is literally the ultimate ideological apparatus which establishes a womans subjected subjectivity(26). There is a very fine line between seeing her as the victim of tradition and seeing her as the agent of positive choice, since the latter also aims at voluntary subjection. Thus highly educated urban women try to side with the woman and as a result end up constructing an image of women different from themselves(27).
Moreover, the anti-sati lobby emphasised the need for legal reforms prohibiting sati and the glorification of sati. This signifies the demand for state intervention and hence it differs little from the nineteenth century imperialist view. In the nineteenth century, legal system was prepared as a means to civilise India whose backwardness was symbolised by sati and thus colonial rule was legitimised(28). At that time, episodes of British people rescuing widows just as they were about to throw themselves into the flames were often reported. It can be said that in 1987 this picture was replaced by that of urban intellectuals civilising savage rural Indians. In this way, the intellectuals, especially the activists in women's organisations in this case, came to be seen as the enemies of "nationalism". They were criticised as women poisoned by western thought who did not understand tradition, and also as agents of the state oppressing the local community. The Rajputs have come to positively accept the position as a minority opposing the modern nation state(29).

6. Sati in the west coast of Sri Lanka
On 5th October 1982, an elder of the Tamil fishing village in the west coast of Sri Lanka where I was staying for my fieldwork died. He was seventy-one years old. A funeral was to take place in his house the next morning. Two barbers were to conduct the funeral. There is a division of labour so that members of the barber caste usually officiate in family rituals such as funerals, which involve impurity, and Brahman priests in those that do not, such as marriage. In the case of funerals too, the Brahman officiates the ritual at the end of mourning when the impurity disappears.
First, the corpse is carried out the house on to a platform set up in the courtyard. The male relatives pour water on the corpse to clean it. Then the women do the same. The barber shaves the old mans beard and moustache. Those who are present rub turmeric and sesame oil on the forehead of the corpse. The women then wash the body with soap. After being washed, the corpse is moved on to a wooden bed placed in the courtyard. There the body is rubbed with sacred ashes and dressed in white waistcloth with gold embroidered borders and white shirt which is the formal wedding attire. The body is sprinkled with perfume and decorated with flowers.
Then the old wife who has been inside the house until then appears in the courtyard led by the hand by another woman. She wears a bright red wedding sari. Not only does this seem to be a mismatch for a funeral atmosphere but also the fact that it is an old woman wearing it makes it appear strange. The wife seems ready to collapse at any moment. She goes round the corpse once and throws the paddy in a mortar on to the corpse. She then sits next the corpse. After receiving the blessings of the women, she lies down on the right hand side of the corpse. A white cloth is placed covering the two bodies and the relatives through paddy over it followed by other participants. When this is finished the wife gets up. Then three women cut and take off her gold necklace called tari which was presented by the husband at the wedding and is the symbol of married women and hence of auspiciousness. A new plain sari is presented to the wife by a male relative. The corpse is then carried to the burial ground. Only men accompany this and the women stay at home.
What I want point out here is that the funeral for the dead is conducted in such a way as to remind us of a wedding and that the farewell to the dead man is also extended to the wife. Both wear a wedding costume and moreover the wife lies next to the corpse. A white cloth is put over them and the unity of the husband and wife is emphasised in a dramatic way. The people bid farewell to both of them. The wife is thought to die along with the husband. She then becomes a widow. The villagers explained this custom in which the wife lies next to the husband as sati.
What kind of an attitude can an anthropologist take towards an alien custom in another country such as sati? First he can clarify the beliefs behind sati. This is what I have presented in the above accounts of the view on ideal women regarding sati and the beliefs regarding blessing and curse. The anthropologist Harlan focused on the word sat, and there is a similar word in Tamil called karpu. This word is also related to concepts such as chastity, heat and protection of husband. If this point is more generalised, much of it overlaps with the discussion above and can become an explanation of the prominent view on women in Hindu society.
In Hindu society, death or funeral has been understood by the concepts impure and inauspicious. Certainly death is impure and the children and the spouse of the deceased have to undergo a certain period of mourning. They accept the impurity caused by death and they are also inauspicious. However, from this point of view, the reason why in Sri Lanka the wife must lie next to the husbands corpse wearing a wedding costume cannot be explained. Wedding dress is a symbol of auspiciousness, and marriage is one of the few life cycle rites that do no cause impurity. Why a wedding dress in a funeral? In order to answer this question we have to take womens sexuality into account.
An ideal life of a woman is that in which her mature sexuality is always controlled by her husband. There are three cases of deviation from this ideal, that is, getting married after menstruation starts, the husband dying before the wife and adultery. Mature sexuality begins with menarche. Therefore an ideal marriage should take place before first menstruation. After first menstruation the woman goes to live with her husband. However, child marriages were legally prohibited by the series of reforms in the colonial period. Today, marriage takes place in most castes after first menstruation, but in cases where first menstruation is ritually celebrated as in South India, the first menstruation ritual itself sometimes assumes a form resembling the marriage rites. Thus first menstruation ritual is also the confirmation and at the same time the control of sexuality, though it is insufficient compared to marriage. In contrast there are cases as those in North India where the first menstruation is ritually completely ignored. I would like to leave the various interpretations of first menstruation and marriage rites for a separate article, but what is important here is that the husbands death should be seen as the end of this kind of control. That is to say, sati is a voluntary proposal of death by women as a means of resistance against the ending of the control of their sexuality that was completed at marriage.
Coming back to the funeral in Sri Lanka, if it were not for the comment that this is a kind of sati we might have found a more romantic and positive meaning for the scene where the widow appears dressed in her wedding dress. We might even have overlooked the glimpse of ritual violence in the cutting off of the tari. But we cannot just interpret this as a symbol of the strength of the love between husband and wife. This is because it is precisely this kind of discourse that has been legitimising sati and it is because of this that it has been the object of criticism. It is also insufficient to ignore the example observed in Sri Lanka as just being symbolic, since by doing so there is the possibility of misjudging the effect and the extent of the violence that is at work in sati. In this way, we can say that the analyses of the funeral in Sri Lanka and the sati in Rajasthan make possible a deeper understanding of each.
Here I want to point out that in the case of the example from Sri Lanka, even if the sati was conducted symbolically, there is no way that the widows status is raised or divinised. It can be said then that the example from the Sri Lankan fishing village is just a poor imitation of sati. This brings to mind the discourse that true sati is only possible for Rajput women. Here the discussion goes back to the previous questions. Why is it the Rajputs who bring about sati? Why Roop Kanwar? To answer these questions, the explanations such as the misery of widows and control of sexuality seem to be much too general. At the same time, these explanations ignore the nationalist view of women which has been historically constructed in India as I have mentioned above. Analytical perspectives such as comparison, investigation of meaning and holistic view which characterise cultural anthropology certainly can be said to contribute to a deeper understanding by linking Sri Lanka and Rajasthan, by not ignoring the symbolism and by taking into consideration the aspect of womens sexuality. By doing so, some light can be shed on the relationship between death and sexuality which has been neglected in previous studies of funerals that emphasis impurity and inauspiciousness. However, it might be said that the anxiety of previous cultural anthropological explanations, which emphasise the whole and meaning, lays in the fact that the better the explanations are the more they end up accepting the violence that is the object of analysis. In the last section, I would like to return to the disputes over sati keeping in mind how far it might be possible to break the link between explanation, understanding and acceptance.

7. From blessing to violence and to pain
Sati is violence in the name of blessing exercised by a goddess choosing one woman out of many widows. It is the family of her in-laws, the village, and the Rajputs who actually exercise the goddess violence. Sati is carried out in the name of the goddess and the community. The chosen widow becomes a goddess, blesses the people in turn and chooses another widow.
In order to deny this belief and to accuse sati as being murder, one would have to deny the power of the goddess and of the community. This is rationalism and secularism which is anti-religious, and also individualism which is anti-community. This is indeed the anti-sati stance. It is said that only the spread of education based on these ideals will truly liberate women. However, in reality, those who are against sati have no choice but to appeal to the state to prohibit sati. From the point of view of the sati defenders, this is state interference in religion and against secularism (guarantee of religious freedom) and in so far as sati is seen as a family problem, it is state intervention in the private life. In other words, the pro-sati lobby also happens to respond using the same logic of secularism and individualism as the anti-sati lobby.
The dilemma against this violence called sati does not end here. For instance, if sati is seen as murder, the people who lit the fire or were directly involved in the sati would indeed be arrested and judged in the name of the law, but the collective nature of the violence would be ignored. Moreover, if the act of murder were emphasised, the violence at the symbolic level would be missed. The everyday discrimination or violence exercised against widows within the home would remain unseen. Furthermore, the power of discourse that distinguishes and differentiates between widows and married women whose husbands are alive would also be overlooked. These points can be said to illustrate the difficulties in criticising violence, especially ritual violence that is legitimised by religion (community) and which is usually only one element in a ritual.
The way to criticise ritual violence such as that represented by sati would be to deny the discourse that it is a divine blessing, that is to say, to make clear that it is not a blessing but violence, and that it does not involve happiness but pain for the one concerned and sorrow for the close relations(30).
For instance, Asahi newspaper dated 13th October 1987 reports Roop Kanwar's sati as follows. The headline was "A young wife in flames embracing her husband who died from illness". The story begins with the sentence 'Kanwar sat on the wood piled up for her husband's cremation and put his head on her lap. She smiled as her body burned in the raging fire and then fell collapsing'. The words "in flames" in the headline and "smiling" are indeed none else than expressions that mystify sati. In Japan, which is far away from India, sati was just a convenient incident to satisfy the curiosity about the mysterious country where anything goes. I will not question their irresponsible attitude here. The denial of mystification or of the negation of pain typically found in the word "smiling", that is to say, the complete denial of the legitimacy of the violence by the divine beings is one of the few ways of going against the divine and communal violence. In other words, we have to ask questions regarding the body of Roop Kanwar which was enveloped in flames and not regarding sati as an institution and Hindu or Rajput nationalism. By doing so we can shift the point of discussion from the problem of rational subject who can make decisions to the problem of embodied subject that feels pain(31).
The dilemma faced by the anti-sati lobby, which emphasises the subjectivity of women, is similar to the dilemma that anthropologists feel when confronted with ritual violence of othercultures such as "female circumcision". This is because anthropologists cannot find a decisive solution as to how they can talk about the victims of violence, how they can resolve the violence itself, and whether they as anthropologists should really say they ought to do so in the first place. This difficulty lies in the fact that they have the beliefs and institutions legitimising violence as their framework of explanation. As long as cultural anthropology respects other cultures from the point of view of cultural relativism, it is not possible to criticise ritual violence. This is more so if it has to do with the identity making of individuals. It is probably necessary to somehow disengage the connection that to understand violence leads to approving and defending it. To focus on pain here would be to point out that the individual is not necessarily totally bound to the systems of meanings or beliefs and social institutions, and does not necessarily become a completely subjected subject. The very fact of a body that responds to and resists pain shows us that there are aspects which cannot be subjectified. Desire and pain rooted in the body undermine the ideologically constructed subject. At that moment, the individual momentarily overcomes the subjected subject. We must not overlook this moment. The focus on pain, however, should not mean considering the individual as an isolated body. Rather, we should make clear the "community of sorrow" which is hidden by the mystification of pain(32). This is not about accusing the relatives of acting out of self-interest and dividing up the pro-sati lobby which emphasises honour. It is rather about looking for people who accept the pain of sati as their own. It is by changing the analytic perspective from blessing to violence and then to pain, from the community of honour to the community of sorrow that the anthropology of violence can overcome many difficulties and open up new frontiers.

ENDNOTES
(1) For details of interpretation of sacred texts and historical sources regarding sati, see V.N. Datta, Sati: Widow Burning in India, Manohar, New Delhi, 1988, Sakuntala Narasimhan, Sati: A Study of Widow Burning in India, Viking, New Delhi, 1990, and Arvind Sharma, Sati: Historical and Phenomenological Essays, Moti Lal Banarsidas, New Delhi, 1988.
(2) For instance, see Vasudha Dalmia-Luderitz, 'Sati' as a Religious Rite; Parliamentary Papers on Widow Immolation, 1821-30, Economic and Political Weekly 27(4), 1992, pp.58-64, Veena Das, Gender Studies, Cross-Cultural Comparison and the Colonial Organization of Knowledge, Berkshire Review 21, 1986, pp.58-76, Hanne Georgeson, Representations of Hindu Women through Some of the Rewritings on Widow-burning, TAJA 3(3), 1992, pp.150-174, Lata Mani, Contentious Traditions: The Debate on SATI in Colonial India, Cultural Critique 7, 1887, pp.119-156, the Female Subject, the Colonial Gaze: Reading Eyewitness Accounts of Widow Burning, in Tejaswini Niranjana, P.Sudir and Vivek Dhareshwar (eds.), Interrogationg Modernity: Culture and Colonialism in India, Seagull Books, Calcuttta, 1993, pp.273-290, Gayatri Chakravorty Spavik, Can the Subaltern Speak? in Cary Nelson and Lawrence Grossberg (eds.), Marxism and the Interpretation of Culture, University of Illinois Press, Urbana, 1988, pp.271-313, Dorothy K Stein, Women to Burn: Suttee as a Normative Institution, Signs 4(2), 1978, pp.253-273. Of course, although the nineteenth century disputes over sati centering in Bengal are not exactly the same as those I deal with in this article, they are not unrelated. In fact, in 1987 the nineteenth century disputes were often mentioned in the newspapers and there the British government was praised. In order to consider the post-colonial condition that India faces, it is necessary to keep the nineteenth century disputes on sati in view.
(3) When I visited Deorala in March 1988, there was a police checkpoint on the road on the way to the village where I had to leave my camera.
(4) Since I deal mainly with English language newspapers and magazines, the opinions are mostly against sati, but I have tried as much as possible to include the pro-sati views found in Hindi language newspapers through translation of them. I would also like to point out that I have not given the sources of the opinions which recur many times in different articles.
(5) Already around a hundred years ago, Durkheim had declared sati to be a collective intentional suicide enforced by a group in his work Suicide [1897]. He said that this occurs in a society where individuals are almost buried in the society and individual personalities are not respected, and when the society neglects life and takes it for granted that individuals give their lives for it without any important reason. See also for R. S. Gandhi, Sati as Altruistic Suicide, Contributions to Asian Studies 10, 1977, pp.141-157. As it will become clear later on, those who criticise sati is faced with the dilemma of having to accept this kind of view of Indian society.
(6) The first newspaper to report this incident was the local paper Rajasthan Patrika (5th September 1987), but it is said that it was treated in much the same way [Sharada Jain, Nirja Misra and Kavita Srivastava, Deorala Episode: Women's Protest in Rajasthan, Economic and Political Weekly 22(45), 1987, pp.1891-1894].
(7) The literacy rate is 70 percent and 40 percent for women. This is very high considering that the average literacy rate in Rajasthan is only 2 percent for women.
(8) There are many speculations regarding the cause of death, such as intestinal inflammation and acute appendicitis. According to one speculation, he committed suicide due to the news he received fifteen days before his death of the failure in his repeated attempt at the entrance examination. The doctor who saw Man Singhs death went into hiding just afterwards and did not respond to police questions.
(9) This is usually done by the eldest son, but the couple had no children. It is said that in the case of sati, a minor is usually chosen in fear of police arrest.
(10) There are several views about the number of spectators. Some reports say that the number was four to five hundred since no one openly said that they saw it in fear of police arrest.
(11) This was because it was feared that the ashes might be misused by magicians (tantri).
(12) Here I have referred in particular to Lindsey Harlan, Perfection and Devotion: Sati Tradition in Rajasthan, in John S. Hawley (ed.), Sati: The Blessing and the Curse:The Burning of Wives in India, Oxford University Press, Oxford, 1994, pp.112-181. See also Kumkum Sangari and Sudesh Vaid, Institutions, Beliefs, Ideologies: Widow Immolation in Contemporary Rajasthan, in Kumari Jayawardena and Malathi de Alwis (eds.), Embodied Violence: Communalising Women's Sexuality in South Asia, Zed Books, London, 1996, pp.240-296, Sudesh Vaid, Politics of Widow Immolation, Seminar 342,pp.20-23, 1988.
(13) Shobha Kavita, Kanchan Shobita and Sharada, Rural Women Speak, Seminar 342, pp.40-44, 1988. In this article many critical voices against sati by the general populace are recorded.
(14) The Rajasthan Land Reform and Resumption of Jagirs Act was established in 1952. Two years later, the first sati after independence took place in Rajasthan.
(15) See Vaid, Politics of Widow Immolation for this point.
(16) For instance see the statement by a member of the communist party given in Times of India (27th September 1987). But at least Advani, the leader of the BJP (Bharatiya Janata Party) which is thought to be representative of Hindu nationalism, criticised sati and stood in opposition against the party members in Rajasthan. Moreover, the World Hindu Association (Viswa Hindu Parishad), which called for the demolition of the mosque regarding the Ayodya problem, also declared a position of opposing all sati since a distinction cannot be drawn between enforced and voluntary sati (Indian Express 17th October 1987). The same goes for the religious sphere, and all were silent except the Shankaracharya of Puri, or took the opposing position as the Arya Samaj did (The Illustrated Weekly of India 2nd May 1988). This exposes the fact that sati is not necessarily all mighty as the symbol of Hindus.
(17) See Partha Chatterjee, Colonialism, Nationalism, and Colonialized Women: the Contest in India, American Ethnologist 16(4), 1989, pp.622-633, for this point.
(18) Sati by low castes are criticised as just being a means for raising status by imitating Rajputs [Harlan, Perfection and Devotion: Sati Tradition in Rajasthan, p.82].
(19) However, there are subtle differences in the same newspaper, for instance in the Rajasthan version there is a strong pro-sati view.
(20) T.N. Kitchlu, Widows in India, Ashish, New Delhi, 1993, gives a detailed account of widows.
(21) This is by no means an unrealistic anxiety. It was thought that young widows like Roop Kanwar would commit a "mistake" at some time or other, even if she were to remain in the in-laws house or go back to her parent's house and bring dishonour to both houses.
(22) Since Roop Kanwars parental house was well off, the dowry included around 460 grams of gold, a fixed bank deposit of 300,000 yen, apart from radio, colour television, refrigerator and so on.
(23) See Madhu Kishwar and Ruth Vanita, the Burning of Roop Kanwar, Race & Class 30(1), 1988, pp 59-67, for this point.
(24) I have not read the report Meena Menon, Geeta Sheshu and Sujata Anandan Trial on Fire: A Report on Roop Kanwars Death, Bombay: Bombay Union of Journalists, 1987, but its details can be estimated from Veena Talwar Oldenburg, the Roop Kanwar Case: Feminist Responses, in John S. Hawley, (ed.), Sati:The Blessing and the Curse.
(25) There were other reports that Roop Kanwar was drugged and she was hardly conscious (Times of India 3rd October 1987). It was also reported that her husband was impotent and she had a lover so there was no way she would have committed sati [Oldenburg, the Roop Kanwar Case: Feminist Responses, p.116].
(26) It is indeed ultimate in the sense that the actual woman involved dies, but subjectification in the sense of Althusser and Foucault fails. Rather, we should say that sati as a spectacle functions as a pre-modern ideological apparatus that teaches other women how they should live or die.
(27) Problems of womens subjectivity regarding sati have been pointed out by DasGender Studies, Dalmia-LuderitzC 'Sati' as a Relitious Rite, Julie Leslie, Suttee or Sati: Victim or Victor? in J. Leslie (ed.), Roles and Rituals for Hindu Women. Pinter, London, 1991, Rajeswari Sunder Rajan, The Subject of Sati: Pain and Death in the Contemporary Discourse on Sati, Yale Journal of Criticism 3(2), 1990, pp.1-28.
(28) See Krishna Iyer, V.R. Maha Sati: Death to Innocents, Main Stream 26(4), 1987, pp.22-26.
(29) For this point, see Kishwar and Vanita, The Burning of Roop Kanwar, pp.64-65, In particular. Veena Das, Critical Events: An Anthropological Peerspective on Contemporary India, Oxford University Press, Delhi, 1995, considers similar problems regarding the glorification of sati.
(30) On the one hand, we should not deny the heartfelt pain of the widow mourning her husband (not necessary due to the misery of the life as a widow) and the possibility of suicide as a reaction to this. The problem then no longer becomes the evaluation of sati, but that of whether suicide is right or wrong. In cases such as female circumcision or female genital mutilation (FGM) where the pain is not denied or hidden and is positively valued as something to overcome, it seems that slightly different arguments to sati need to be made. For this point see C J. Walley, Searching for "Voices": Feminism, Anthropology, and the Global Debate over Female Genital Operations, Cultural Anthropology 12(3), 1997, pp.405-438.
(31) Rajan, The Subject of Sati, focuses on pain from a similar point of view. Oldenburg, the Roop Kanwar Case: Feminist Responses, p.125, takes this up and points out that the widows body as a subject of desire is also hidden in the representation of sati and argues that the problem should focus on the widow as an embodied being. Of course, we should not forget the fact that the discourses about pain in social science, law and medicine transform pain as it is and disengage it from the people who experience it [Das, Critical Events, p. 175, also, Language and Body: Transactions in the Construction of Pain, Daedalus 125(1), 1996, pp.67-91, Arthur W. Frank, The Wounded Storyteller: Body, Illness, and Ethics, The University of Chicago Press, Chicago, 1995, pp.178-179, E.ScarryCThe Body in Pain: the Making and Unmaking of the World, Oxford University Press, Oxford, 1985].
(32) This corresponds to the concept of "a collective subjectivity of agents" by Ania Loomba, Dead Women Tell no Tales: Issues of Female Subjectivity, Subaltern Agency and Tradition in Colonial and Post-Colonial Writings on Widow Immolation in India, History Workshop 36, 1993, pp.209-227, which was triggered off by Rajan, The Subject of Sati. Das, Critical Events, pp.194-196, proposes anthropology of pain and emphasises the significance of research on this kind of community.

Masakazu TANAKA, Ph.D.(London), Associate Professor(Institute for Research in Humanities, Kyoto University).