This study examines three issues related to domestic violence in India on the basis of data from the third National Family Health Survey. It shows that a significant proportion of women, regardless of their socioeconomic background, accept power differentials based on sex and that men have a right to discipline them. It also reveals that men from violent homes are significantly more likely to use violence against their wives. Most women do not seek help for the violence suffered, and the few who do, tend to steer clear of social service organisations and the authorities.
NATIONAL FAMILY HEALTH SURVEY-3november 29, 2008 EPW Economic & Political Weekly60Violence against Women in India: Is Empowerment a Protective Factor?Leela VisariaSocial history and novels in many Indian languages record violence against married women in India, mainly perpe-trated by their husbands. Yet, only in the last two decades has a systematic effort been made to estimate the magnitude of violence, its determinants and reasons, the forms inwhichit is manifested, and its health, social, legal and economic consequences. IntroductionSurvey-based studies have indicated that anywhere from 35 to 75% of women in India face verbal, physical, or sexual violence from their partners or other men known to them (see Jejeebhoy 1998; Mahajan 1990; Karlekar 1998; Jain et al 2004; Visaria 2000). Qualitative in-depth studies have thrown light on a range of issues such as women’s support-seeking behaviour, intergen-erational effects, the culture of silence, and the adherence to social norms that encourage tolerating, accepting, and even rationalising domestic violence for the sake of preserving family honour (Hassan 1995; Miller 1992; Jaisingh 1995; Koenig et al 2006). However, most of these studies were conducted with small samples and the findings could be generalised even for the states in which they were conducted. Also, very few studies have been carried out to examine these issues from the perspective of the perpetrators of violence. To overcome this limitation, the second National Family Health Survey (NFHS-2), conducted in 1998-99, took the bold step of nationally canvassing a few questions related to domestic violence, essentially to assess whether women in a large survey would respond to them.1 The questions were rather general and tried to measure the prevalence of violence and understand the situations in which ever-married women justified wife-beating. The respondents were read out six situations where wives trans-gressed their traditionally accepted roles or social norms. Women’s responses were sought on whether their husbands were justified in beating them if they deviated from their perceived “duties”.2 The success with canvassing these relatively sensitive questions emboldened the coordinators and advisers of the third NFHS (2005-06) to canvass an entire module on domestic violence with 25 key questions besides the ones on wife-beating (IIPS and Macro International 2007). InNFHS-3, two more situations were added and one dropped in the wife-beating question. The two new situations were arguing with the husband and refusing to have sexual intercourse with him. The situation of the woman’s family not providing the expected money, jewellery or other goods (implying dowry) was dropped This study examines three issues related to domestic violence in India on the basis of data from the third National Family Health Survey. It shows that a significant proportion of women, regardless of their socio-economic background, accept power differentials based on sex and that men have a right to discipline them. It also reveals that men from violent homes are significantly more likely to use violence against their wives. Most women do not seek help for the violence suffered, and the few who do, tend to steer clear of social service organisations and the authorities.Leela Visaria (lvisaria@gmail.com) is at the Gujarat Institute of Development Research, Ahmedabad.
NATIONAL FAMILY HEALTH SURVEY-3Economic & Political Weekly EPW november 29, 200861because less than 7% of the women inNFHS-2 cited it as a reason for violence. The same set of questions on attitudes towards wife-beating was asked to both women and men inNFHS-3. Interestingly, nearly the same proportion of women (54%) and men (51%) agreed that wife-beating was justified in at least one of the situa-tions. Also, the result was similar when the background charac-teristics of the respondents were considered. Among both men and women, a greater proportion of those who were illiterate, poor, young, and those in rural areas who were divorced or deserted justified wife-beating more than the others. It suggests that there is good agreement between men and women on the norms that govern married life and that if women fail to observe the norms they would bring violence upon themselves. The questions in the domestic violence module in the NFHS-3 sought information on the various forms of violence (physical, sexual or emotional) and the ways in which they were inflicted on all women, regardless of their marital status. Questions about the drinking habits of the partner (to understand its relationship to violence), spousal violence, help-seeking behaviour, the physi-cal consequences of spousal violence, and knowledge of parental violence were asked only to ever-married women. In addition, ever-married women were asked whether marital control was exercised in six specific situations. Some of the questions related to lifetime violence and some had a reference point of one-year. The responses to most of these questions were examined against the background characteristics of the women, and the data help understand some of the barriers to empowerment of women in the country. The results of this enquiry are presented in a separate chapter in the NFHS-3 report.This paper largely examines data from the NFHS-3. Rather than examining all the data collected on violence, I have examined findings on three issues which have not often been empirically studied in depth in India.3 The three issues are: degree of marital control exercised by husbands and other empowerment indicators of women and their relationship to violence – the association between knowledge of parental violence and experience and acceptance of spousal violence; and the extent to which women seek or expect to receive help to stop spousal violence. TheNFHS-2 reported that 21% of ever-married women experi-enced violence at some time since the age of 15 while the NFHS-3 says that 33.5% of all women aged 15 to 49 have done so.4 However, one cannot, on the basis of the increase in the preva-lence of violence between the two NFHS surveys, surmise that domestic violence has been increasing in India on the basis of these figures alone. The NFHS-2 included only a few questions to ascertain whether it was possible to collect data on violence in a large survey. There was also a strong likelihood of substantial under-reporting because women would have been very reluctant to respond positively to experiencing violence if the questions were asked in the presence of family members and others. The investigators were instructed not to ask the questions on violence, which were part of the ever-married women’s questionnaire, in the presence of others. However, it is difficult to know how far they succeeded in adhering to this instruction. In theNFHS-3, a separate module was prepared and canvassed only to one woman in each household, not to all the eligible women if there were more than one. More importantly, the respondents were clearly instructed to answer the violence module only if they were ensured complete privacy.5 Although estimates of lifetime physi-cal violence are available for all the states as well as the background characteristics of women respondents in the NFHS-2 andNFHS-3, it would not be prudent to discern a time trend given the variations in the mode of data collection. I shall, therefore, only examine the data available from the NFHS-3, the most recent survey.Aspects of Violence against WomenTo contextualise the three issues that are examined in the follow-ing sections, I briefly present some of the salient findings on the extent of violence reported by women and the differentials in accordance with their background characteristics and the region in which they live. Compared to women in the 15 to 24 age group, a greater proportion of older women experienced lifetime violence. A higher percentage of rural women (36%) were subjected to violence than women living in urban areas (28%), and a significantly greater proportion of women with little or no education experienced violence compared to their educated sisters. Only 14% of women with 12 or more years of schooling reported experiencing violence while the figure for illiterate women was 44%. Violence was reported by a higher percentage of currently married women (37.4%) than never married women (16.1%). But 66% of divorced, separated or deserted women reported having experienced physical violence. Violence was much more prevalent among women from the scheduled castes and tribes (39-42%) than among those from the higher castes (27%). Violence was also inversely related to the wealth index. Violence against women varied hugely among the states. More than 40% of women aged 15 to 49 reported having experienced physical or sexual violence in the relatively backward states of Rajasthan, Madhya Pradesh, and Bihar. Interestingly, among the big states, the percentage was only a tad lower in West Bengal, Uttar Pradesh and Tamil Nadu. On the other hand, less than 20% of women reported experiencing violence in states like Himachal Pradesh, Kerala, and Karnataka. It appears that rather than the economic development of a region, it is the attitude towards women, societal norms and perceptions about theirworthand position in the household, and men’s notions of self-esteem that influence husbands’ behaviour, for better or worse. In spite of these differentials, it is important to note that one in five women from the wealthiest group and one in seven women with educa-tion of 12 or more years reported being subjected to violence within the home, almost always by the spouse. A relatively high percentage of Tamil women reported violence compared to those from many other states and this merits atten-tion. An in-depth study undertaken in the slums of Tamil Nadu’s capital, Chennai, reported that the men held that women should be disciplined. They required their wives to be chaste, submissive, respectful and accepting of their imperfections (Go et al 2003). Wife-beating was condoned to ensure that women behave themselves and stay under the control of men. According to
NATIONAL FAMILY HEALTH SURVEY-3november 29, 2008 EPW Economic & Political Weekly62NFHS-2, while 21% of ever-married women in the country as a whole said they had experienced lifetime violence, the percentage for Tamil Nadu was 40, the highest in the country (IIPS and ORC-Macro, Tamil Nadu, 2001). According to theNFHS-3, almost the same percentage (39%) of Tamil women reported experienc-ing physical or sexual violence. Further, acceptance of violence as a justifiable behaviour was also greater in Tamil Nadu. According to theNFHS-2, in the country as a whole, about 56% of ever-married women said their husbands were justified in beating them if they failed to perform certain duties such as showing respect to their in-laws or attend-ing to their children and household. In Tamil Nadu, 72% of women accepted violence as a justifiable act, thereby testifying that gender inequality was quite pervasive there (Kishor and Gupta 2004). According toNFHS-3, nearly two in three Tamil women agreed that in certain situations husbands were justified in beating their wives. An in-depth exploration of the acceptance of domestic violence by the women of Tamil Nadu would help understand the cultural and social norms that govern conjugal relationships in Tamil society. However, there are almost no such behavioural guidelines or restrictions apply to men in any Indian society. Risk Factors Husbands tend to exercise control over their wives by clearly in-dicating how they should behave. InNFHS-3, information was sought from ever-married women on six specific situations: whether the husband became jealous or angry if they talked to other men; whether the husband accused them of being unfaith-ful; whether the husband would not permit them to meet their female friends; whether the husband tried to limit contact with their natal family; whether the husband insisted on knowing where they are at all times; and whether the husband did not trust them with money. These situations reflected different di-mensions of women’s lives, ranging from economic independence and mobility to freedom to interact with friends and men known to them without arousing suspicion. While many women may not personally approve of such controlling behaviour, their accept-ance of it or inability to reject it indicate they are not empowered even within their marital home. The responses of women, classified by their background characteristics, are presented in Table 1. A little over a quarter of ever-married women reported that their husbands (or ex- husbands if the respondents were not currently married) became angry or jealous if they talked to other men. Some micro studies have also reported this controlling behaviour to be fairly widespread in many states of India (Jain et al 2004; Visaria 2000). Husbands show their anger or displeasure even when women reportedly talk to their brothers, male cousins or other male relatives from their natal villages, or neighbourhoods in the case of urban areas. A married woman having a male friend or visitor almost amounts to blasphemy and it becomes the subject of gossip not only in rural areas but also in many urban settings. A greater proportion of younger, recently married women in rural areas, with little or no education, who belonged to poor house-holds, and those who were divorced or separated, faced the wrath of their husbands (or ex-husbands) on this count than better educated older women or urban women. The logic seems to be that women’s behaviour has to be checked when they are young and newly married so that they learn to behave (that is, not be familiar or friendly with other men) according to social or family norms. The anger, jealousy or suspicion of husbands sometimes manifests itself in accusations that the wife is being unfaithful or having illicit relations with other men. Nearly 9% of women reported that they were frequently accused of being unfaithful. Again, the differentials were in the same direction as with jealousy. The husbands of nearly 12% of women insisted that they know where their wives were at all times. This desire to know their wives’ every movement largely stems from wanting to check that they do not see other men or talk about family problemsto others. Less educated younger women, those living in rural areas, and those divorced, separated or remarried from the lower wealth group faced this more than women belonging to other categories.6 Further, nearly 16% of women reported that they were not permitted to meet even female friends. In areas or communities where exogamy is practised and marriages are arranged by parents or other elders, women do not always have many acquaintances from their places of origin or areas near their marital home. The restriction enforced on women interacting with other women known to them is a very harsh controlling measure. Almost 10% of women also reported that their husbands tried to limit their contacts with members of their natal families. This is manifested by not allowing women to visit their natal family, except when it is absolutely essential, or not welcoming members of their family, or showing displeasure when they visit. Table 1: Percentage of Ever-Married Women Aged 15-49 Whose Husbands Exercised Marital Control, according to Select Background Characteristics( 2005-06) Percentage of Women Whose Husband Background Is Jealous Accuses Her Does Not Tries Insists on Does Displays Displays Characteristics If She Talks of Being Permit Her Limiting Knowing Not Trust Three or None of the to Other Men Unfaithful to Meet Her Contact Where Her with More Specific Female with Natal She Is All Any Specific Behaviours Friends Family the TimeMoney Behaviours Age 15-19 33.6 11.7 19.3 12.3 16.1 20.9 17.2 49.5 20-24 27.7 8.2 15.9 10.3 12.7 18.5 12.7 56.3 25-29 27.1 8.7 16.1 9.8 12.7 18.0 12.4 55.9 30-39 25.8 8.5 15.9 10.0 11.2 18.8 12.0 57.5 40-49 23.5 7.7 14.9 8.7 10.4 17.1 10.4 59.8Residence Urban 20.5 6.4 14.3 8.0 9.3 17.1 9.6 63.7 Rural 29.0 9.4 16.6 10.7 13.0 18.9 13.3 54.0Education Illiterate 32.3 1.08 16.5 11.4 13.4 19.3 14.5 51.5 < 5 years 28.3 9.7 19.7 11.6 14.8 21.2 14.8 53.112+ years9.6 2.6 12.4 4.5 6.2 14.05.0 73.8Caste/tribe SC 29.7 10.5 17.6 10.7 12.9 20.2 14.1 53.7 ST 31.3 12.0 17.6 11.0 15.3 18.9 16.0 52.7 OBC 27.2 8.2 13.6 9.8 11.1 16.1 11.5 58.3 Other 22.2 6.6 17.4 9.1 11.4 19.9 10.9 58.3Wealth index Lowest 33.9 12.417.212.815.419.816.049.3 Middle 28.8 10.0 16.0 10.5 12.3 17.9 13.0 55.6 Highest 14.8 3.0 14.0 5.8 7.1 16.6 6.8 68.1 All 26.4 8.5 15.9 9.9 11.9 18.3 12.2 56.9Not all educational or wealth index categories are presented in the table. Source: IIPS and Macro International 2007, Vol II, pp 505-06.
NATIONAL FAMILY HEALTH SURVEY-3Economic & Political Weekly EPW november 29, 200863When women are married to men from the same village or town, they may experience greater freedom or find ways to visit or interact with their family members without being noticed. But when distance is a factor, this restricting or controlling behaviour tends to have a detrimental effect on women, mainly because they have no opportunities to share their problems with their family members or near ones. Not granting even a modicum of economic independence, especially to women who have no other source of income, is another controlling behaviour. It leads to husbands saying women cannot be trusted with money, implying that they do not know how to spend money judiciously. A little over 18% of ever-married women indicated that their husbands do not trust them with money. This controlling behaviour is expressed by asking women to explain how every rupee is spent and reprimanding them if something is spent on what husbands consider unnecessary. Background characteristics hardly made a difference in the case of this controlling behaviour, which stems from the general notion that women are not careful about what they spend money on. The NFHS-3 estimates that 12% of women reported three or more controlling behaviours by their husbands. The differences when background characteristics were taken into account were not very significant, except that women from poor households faced more controlling behaviour than those belonging to better-off households. Instead of grouping any three types of controlling behaviour, it would be interesting to group the three types of behaviour that prompt husbands to suspect and mistrust their wives when they deal with other men, even their male kin. In a sense, such behaviour undermines the very basis of a marital relationship. Interestingly, the differences between women belonging to various socio-economic groups were very small when it came to restrictions on meeting female friends and handling money. The former controlling behaviour very likely stems from the fear that women will share news about family matters that husbands or in-laws do not want divulged to outsiders. The underlying fear is that women may do so until their loyalty to the family of their in-laws is established. So, young women, even educated ones, are not trusted. Table 1 also shows that 57% of women reported that their husbands do not display any of such specific controlling behav-iour, implying that they are trusted by their spouses. Conversely, 43% of women reported that their husbands show at least one type of controlling behaviour, and they were asked their opinion about this. As expected, the extent of trust was greater among older women (some of whom may have experienced greater control over their behaviour when they were younger; with the passing of time, they gain the trust of their husbands and in-laws), better educated women and among those belonging to better-off households. The extent of violence experienced by women was also examined on the basis of some characteristics of the husband and selected indicators of women’s empowerment. The data presented in Table 2 show that a greater proportion of husbands who are either illiterate or have little education inflict violence (physical, sexual or emotional) on their wives than husbands who are better educated. All the same, women’s education has a much stronger association with violence than men’s education. One in four men with 12 or more years of schooling used violence against their wives but only 15% of women with the same level of education reported being subject to violence by their spouses. It is only education beyond 12 years of schooling that appears to empower women and act as a protective factor. Also, as evident in Table 2, women who have the same level of education as their husbands are least likely to suffer either physical or sexual violence compared to those who are illiterate or have less education than their husbands. Drinking alcohol is very significantly associated with both physical and sexual violence. Seven out of 10 men who get drunk subjected their wives to violence as against three out of 10 men who did not drink. Also, a quarter of men who get drunk inflicted sexual violence on their wives. The combination of sexual desire and alcohol increased women’s risk to violence if she refused sex. A study conducted in south India indicated that the risk of wife abuse increases significantly with alcohol consumption by the husband (Rao 1997). Another study carried out in Karnataka reported that independent of caste and economic status, alcohol consumption by husbands was found to be significantly associa-ted with violence (Krishnan 2005). More than moderate consump-tion of alcohol by men definitely increases the chances of violence against women. In addition, 81% of men who displayed five or all of the six controlling behaviours inflicted physical or sexual violence on their wives. Two out of five men who exercised a high degree of marital control also sexually violated their spouses. The Table 2: Percentage of Ever-Married Women Aged 15-49 Whose Husbands Have Subjected Them to Physical or Sexual Violence, according to Select Husbands’ Characteristics, and Empowerment Indicators(2005-06)Characteristics PhysicalViolenceSexual Violence Physical or Sexual ViolenceHusband’s education Illiterate 45.2 12.8 47.2 < 5 years 43.8 13.0 45.7 8-9years 34.2 10.4 36.6 12 + years 18.9 4.9 20.8Husband’s alcohol consumption Does not drink 28.0 7.9 30.3 Gets drunk sometimes 46.8 11.6 48.5 Gets drunk very often 67.6 23.6 68.6 Spousal education difference Husband better educated 34.1 9.7 36.4 Wife better educated 29.7 9.0 31.9 Both equally educated 21.4 6.0 23.3 Bothilliterate 46.0 12.8 47.8Number of marital control behaviours displayed by husband None 23.9 5.0 25.6 1-2 44.1 12.7 46.8 3-4 60.2 23.2 63.0 5-6 77.6 39.1 80.9Number of decisions in which women participate None 32.9 11.6 36.1 1-2 35.9 11.0 38.3 3-4 34.4 8.4 36.1Number of reasons for which wife-beating is justified None 28.2 8.4 30.4 3-4 42.0 11.6 47.7 All7 41.4 11.9 42.8 All ever-married women 35.1 10.0 37.2Source: IIPS and Macro International 2007, Vol II, pp 511-12.
NATIONAL FAMILY HEALTH SURVEY-3november 29, 2008 EPW Economic & Political Weekly64controlling behaviours stem from a lack of trust in women and leads to violence against them.To understand women’s participation in decision-making in the household, the NFHS-3 asked women whether they participatedin decisions related to their own health, major household purchases, purchases for daily household needs, and on visits to their family and relatives. If women did not participate in any of these decisions, they received a score of zero. Those who participated in one or two decisions were viewed as moderately empowered and those who had a voice in three or all four decisions were viewed as highly empowered. As evident in Table 2, empowerment has no clear relationship to prevalenceof violence. This belies the expectation that women who participate in house-hold decisions, and therefore have egalitarian gender-role attitudes, are less likely to experience violence. On the other hand, 42-44% of women who indicated that wife beating was justified in any one of the six situations were themselves recipients of physical or sexual violence compared to 30% of those who said none of the situations excused violence. Overall, well-educated women and women in marital relationships where the husbands did not display controlling behaviours were the ones most likely to escape violence. Intergenerational EffectIn theNFHS-3, ever-married women were asked a question on whether their mothers were beaten by their fathers. Responses to this reflected the extent to which young girls who witness parental violence, knowingly or unknowingly, accept violence as a part of their own married life. The data presented in Table 3 show that two-thirds of women who knew that their mothers were beaten by their fathers experienced some form of violence at the hands of their husbands. For nearly 60%, the violence was physical or sexual. The probability of children who have witnessed parental violence inflicting the same on their spouses when they grow up is quite high and a cause for concern. A third of women who said they had not seen any parental violence also reported being subjected to violence by their husbands. It is likely that some women who said that they were not aware of, or did not know about, parental violence were reluctant to disclose what happened between their parents. A fairly large study carried out to understand the behaviour of men in Uttar Pradesh reported that husbands who had witnessed their fathers beating their mothers as children were 4.7 times more likely to beat their own wives and three times more likely to sexually coerce them than men who had not witnessed such violence (Koenig et al 2006). Martin et al (2002) showed that witnessing violence between one’s parents while growing up is an important risk factor for the perpetra-tion of violence on one’s partner in adulthood. Compared to men raised in non-violent homes, men from violent homes were significantly more likely to believe in the right to control their wives and to physically and sexually abuse them. The study also demonstrated that non-violence in the earlier generation was strongly predictive of non-violence in the second generation.Although theNFHS-3 included only one question on the inter-generational effect of domestic violence, this is an area which needs much more exploration in-depth. The effect of witnessing violence on the minds of children, the internalisation of the prevailing norms related to violence, the subsequent behaviour and the rationalisation of that behaviour all need to be investi-gated while addressing the issue of violence and ways to break the cycle of violence. Help-Seeking Behaviour In the NFHS-3, all women who reported physical or sexual violence were asked a number of questions on whether they sought help to try and end the violence. Women who said that they did seek help were asked from whom they sought it. Also, those who reported that they did not seek any help were also asked questions on whether they took anyone into confidence and shared their plight with them. Table 4 shows some of the data classified according to the background characteristics of the women. Only about one in four women (23.8%) sought help to end the violence they experi-enced. Two out of three women neither sought help nor didtheytellanyone (family members or friends) about experi-encing violence. What is quite striking is that there are virtually no differentials in telling others about the violence, or seeking help from someone, whatever the background characteristics. Neither education nor family wealth act as protective factors in this Table 3: Percentage of Ever-Married Women Aged 15-49, Who Reported that Their Father Beat Their Mother, Experiencing Violence by Their Husbands(2005-06)Respondent’s Father Emotional Physical Sexual Any of the Beat Her Mother Violence Violence Violence Three FormsYes 28.5 57.4 59.6 62.2No 12.4 28.3 30.3 32.8Do not know 18.0 46.5 49.7 52.1Source: IIPS and Macro International 2007, Vol II, p 510.Table 4: Per Cent Distribution of Women Aged 15-49 Years Who Told Someone about Experiencing Violence and Sought Help to End It, according to Background Characteristics(2005-06)Background Characteristics Never Told Anyone % Who Told Someone Sought Help from Any SourceResidence Urban 65.2 8.6 23.6 Rural 66.4 7.3 23.9Education Illiterate 66.5 7.8 23.7 < 5 Years 65.9 7.8 24.0 8-9 Years 66.9 8.2 22.3 12 or more years 69.0 6.8 20.0Caste/tribe SC 64.8 7.0 26.1 ST 65.4 9.5 23.2 OBC 65.0 7.8 24.6 Other 69.2 7.3 20.9Wealth index Lowest 67.1 8.0 23.2 Middle 66.1 7.1 24.8 Highest 67.8 8.1 20.4Type of violence Physicalonly 68.1 7.5 21.7 Sexualonly 85.3 4.1 7.7 Both physical and sexual 52.7 9.5 36.7 All 66.1 7.7 23.8Source: IIPS and Macro International 2007, Vol II, pp 522-23.
NATIONAL FAMILY HEALTH SURVEY-3Economic & Political Weekly EPW november 29, 200865regard. In fact, better educated women and those belonging to families with a better economic status were more unlikely to share their experience of violence with others. Women who experienced sexual violence were even more reticent about talking to others about it or seeking help from anyone. The silence surrounding violence in general and sexual violence in particular has been reported in other micro and in-depth studies (Visaria 2000 and Visaria 2002). This has to be understood in the context of women trying to preserve the family’s honour by not disclosing something that happened within the home, and also in the context of the sense of shame associated with being abused by someone known to them and with whom they share intimate or conjugal relations. In such a situation, from whom do battered women seek support? In theNFHS-3, ever-married women were asked this question. The majority of women who experienced violence and sought help reported that they did so from their natal family; 71% turned to their parents and other family members for support. Nearly 30% sought help from husbands’ families.7 Neighbours were resorted to by 15% and 9% turned to friends. Many a time, neighbours witness the violence and occasionally also get involved in an attempt to ease the situation. Hardly any women opted to report cases of violence to formal organisations or the authorities, presumably because they feared being ostracised and shamed by the communities in which they live. The fear that they themselves will be blamed for provoking husbands to use violence is all too real. It is the reality of Indian society that women who summon the courage to challenge their abusers in a court of law or seek the support of social service organisations have to be prepared to face a long and humiliating battle with little sympa-thy from the authorities or family members, and even the media. Interviewing women survivors of abuse in New Delhi, Prasad (1999) demonstrated that the legal system and procedures designed to enhance women’s access to the law actually inhibited it, and that the state as such showed a tolerance of domestic and sexual violence.ConclusionsIn this paper, I have largely examined three issues related to domestic violence in India. They are the degree of marital control exercised by husbands, and other empowerment indicators of women and their relationship to violence, the association between knowledge of parental violence and experience and acceptance of spousal violence, and the extent to which women seek or expect to receive help to stop spousal violence. The underlying question is whether women’s empowerment is a protective factor that safeguards them against violence. The findings show that a significant proportion of women, regardless of socio-economic background, subscribe to power differentials based on sex and accept that men have a right to discipline them, especially when they fail to fulfil gender- specific duties such as taking care of the house and children or cooking food on time in a manner that pleases the husband. Further, women who are beaten or otherwise physically abused tend to justify their husbands’ behaviour, as a way of rational-isingthe treatment meted out to them.Admittedly, the subordination of women and the dominance of husbands when they are perceived to be transgressing from their wifely duties is not unique to India and cuts across cultures and nations. Nonetheless, the experience of violence, or even the threat of violence, and controlling behaviour by their spouses tends to lower women’s self-esteem, instil fear in them and further lower their ability to fulfil daily tasks to the satisfaction of the members of the family. The controlling behaviour that prompts husbands to suspect the moral character of their wives and mistrust their dealings with other men, including their male kin, undermines the very basis on which a marital relationship rests.Witnessing violence between one’s parents while growing up has been found to be an important risk factor for the perpetra-tion of violence on partners in adulthood. Men from violent homes are significantly more likely to believe in husbands’ rights to control their wives, and to be physically and sexually abusive towards them. The internalisation of the prevailing norms related to violence, and the subsequent behaviour and rationalisation of that behaviour need to be examined while addressing the issue of violence and the means to break the cycle of violence.As in other surveys, in the NFHS-3 hardly any women reported that they sought redress or support from formal organisations or the authorities to deal with the violence meted out to them for fear of being ostracised and shamed by the communities in whichthey live. The fear that they themselves will be blamed for provoking men to use violence against them looms large. In the absence of supportive shelters or other avenues, it is very difficult for battered women in India to gather the courage to challenge their abusers in a court of law or seek the support of the few social service organisations that exist. The humiliation that those who approach the judiciary face because of long court battles, and having to deal with an insensitive police and others with little sympathy unnerve most women. They choose to suffer in silence in their homes, which become dysfunctional.Even education does not empower women to enter the public arena for support. Better educated women or those belonging to better-off families who experience violence are least likely to share their experiences or seek support from others. This needs to be understood in the context of a culture of silence where women try to not disclose what happens within the home environment. Equally important is the sense of shame associ-ated with being abused by someone known to them and with whom they share intimate or conjugal relations. Even when physically injured, women remain silent and suffer it alone. Further, societal norms that tolerate and accept violence are widely prevalent in Indian society and an adherence to them prevents women from seeking care. Finally, an in-depth exami-nation of the widespread acceptance of spousal violence among women of Tamil Nadu would help understand the prevalent cultural and social norms that govern conjugal relationships in Tamil society.
NATIONAL FAMILY HEALTH SURVEY-3november 29, 2008 EPW Economic & Political Weekly66Notes 1 The question in the NFHS-2 which asked whether a woman had been beaten or physically mistreated ever did not offer any definition of what consti-tuted violence. So a woman for whom an occasional slap from her husband qualified as routine or normal behaviour was not likely to regard it as an act of physical violence and report it as such.2 The situations in the NFHS-2 wife-beating question were: the husband suspecting his wife of being unfaithful, the wife’s family not giving the expected amount, jewellery or other goods, the wife showing disrespect to her in-laws, the wife going out without telling the husband, the wife neglectingthehouseor children, and the wife not cooking food properly (IIPS and ORC Macro 2000, p 72). 3 The prevalence of life-time physical violence against Indian women has been estimated by a number of research studies. Similarly, the forms of violence and the reasons for them have also been explored at some length. Some relatively large studies have also examined these against the backgrounds of the respondents to under-stand whether there are differentials in violence experienced by women by place of residence, age, education, employment status or social group.4 Unlike the NFHS-2, which collected information from ever-married women, the NFHS-3 asked some of the violence-related questions to all women, including single women aged 15 to 49 years. 5 For the first time, in the NFHS-3, men were also interviewed and questions were asked on whether they were justified in beating their wives in certain situations, and whether they had the right to get angry or reprimand their wives if they refused sex. These data are available in the NFHS-3 report but are not analysedinthispaper.6 One may argue that one of the reasons for divorce or separation is that women behave freely with other men and thereby invite the wrath or suspi-cion of their husbands. One can equally argue that it is the controlling and suspicious nature of husbands that stifle women’s freedom which results in divorce/separation. A very sensitively conducted in-depth study would help unravel such complexities in spousal relationships. 7 Women reported seeking help from multiple sources and so the figure can exceed 100.ReferencesGo, V F et al (2003): “When HIV-Prevention Messages and Gender Norms Clash: The Impact of Domestic Violence on Women’s HIV Risk in Slums of Chennai, India”, Aids and Behaviour, Vol 7, No 3, pp 263-72.Hassan, Yasmeen (1995):The Haven Becomes Hell – A Study of Domestic Violence in Pakistan, Special Bulletin (Lahore: Shirkat Gah). IIPS and Macro International (2000):National Family Health Survey (NFHS-2), 1998-99: India (Mumbai: International Institute for Population Sciences). – (2001): National Family Health Survey (NFHS-2), India, 1998-99: Tamil Nadu (Mumbai: Inter-national Institute for Population Sciences). – (2007): National Family Health Survey (NFHS-3), 2005-06, India,Vol I (Mumbai: International Institute for Population Sciences). – (2007): National Family Health Survey (NFHS-3), 2005-06, India, Vol II (Mumbai: International Institute for Population Sciences).Jaisingh, I (1995): “Violence against Women: The Indian Perspective” in J Peters and A Wolper (eds), Women’s Rights, Human Rights (New York: Routledge).Jejeebhoy, Shireen (1998): “Wife Beating in Rural India: A Husband’s Right? Evidence from Survey Data”,Economic & Political Weekly, Vol 33, No 15, 11 April, pp 855-62.Karlekar, M (1998): “Domestic Violence”,Economic & Political Weekly, Vol 33, No 27. Kishor, S and K Gupta (2004): “Women’s Empower-ment in India and Its States: Evidence from the NFHS”,Economic & Political Weekly, Vol 39, No 6, pp 694-712. Koenig, M A et al (2006): “Individual and Contextual Determinants of Domestic Violence in North India”,American Journal of Public Health, Vol 96, pp 132-38.Krishnan, S (2005): “Gender, Caste, and Economic Inequalities and Marital Violence in Rural South India”, Health Care for Women International, Vol 26, No 1, pp 87-99. Mahajan, A (1990): “Instigators of Wife Battering” in Sushama Sood (ed), Violence against Women (Jaipur: Arihant Publishers), pp 1-10.Martin, S L et al (2002): “Domestic Violence across Generations: Findings from Northern India”, International Journal of Epidemiology, Vol 31, No 3, pp 560-72. Miller, Barbara D (1992): “Wife-beating in India: Variations on a Theme” in D A Counts, J K Brown and J C Campbell (eds),Sanctions and Sanctuary: Cultural Perspectives on the Beating of Wives (Boulder: Westview Press), pp 173-84.Prasad, S (1999): “Medicolegal Response to Violence against Women in India”, Violence against Women, Vol 5, No 5, pp 478-506.Rao, Vijayendra (1997): “Wife-beating in Rural South India: A Qualitative and Econometric Analysis”, Social Science and Medicine, Vol 44, No 8, pp 1169-80.Visaria, L (2000): “Violence against Women: A Field Study”, Economic & Political Weekly, Vol 35, No 20, May 13, pp 1742–51. – (2002): “Many a Slip between Slap and Lip”, The Little Magazine, Vol 3, Issue 4, pp 40-43.Hind Swaraj Centenary International SeminarSocial Development and Human Civilisation in the 21st Century12-14 February 2009, India International Centre, New DelhiJointly organised by Council for Social Development, India International Centre, DCRC, Delhi University, CPS, Jawaharlal Nehru University, Nelson Mandela Centre, Jamia Milia Islamia, ISHS, University of South Africa, REGGEN, Federal Flomenance University of Brazil & Global Studies, University of California, Santa BarbaraCall for PapersSeveral institutions have come together to organise a major international seminar on the occasion of the centenary of Gandhi’s Hind Swaraj. The objective is to take Hind Swaraj as a vantage point and engage ourselves on some of the fundamental questions our age and try to make a contribution to basic thinking on those issues. The following panels have been proposed to reflect on challenges of the 21st century from the perspective of people of Asia, Africa and Latin America. • Development Theory •Peace • World Order • Civilization and Culture • Self Determination, Liberation and Freedom • Science, Technology and Swaraj • Human Rights and Social Justice • Religion and Social Imagination • Feminism • Swaraj, State and the Marginalised • Field reports on Gram Swaraj experiments • Hind Swaraj as a classical text in world literature Abstracts of 250 words are invited from scholars and activists for the above panels by 15 December 2008. Authors of selected papers will be intimated by 31 December 2008. The last date for the submission of full length papers of 5000 to 6000 words is 31 January 2009. Local hospitality will be provided to all selected participants. Limited travel support is available for international participants and for scholars and activists within India. Contact: Seminar Convener, Council for Social Development53 Lodhi Estate, New Delhi-110003, IndiaPhone: 91-11-24615383, 24611700, 24616061 (Fax) Emails: swaraj@csdindia.org, dr_mohanty@yahoo.com, csdnd@del2.vsnl.net.in
NATIONAL FAMILY HEALTH SURVEY-3november 29, 2008 EPW Economic & Political Weekly60Violence against Women in India: Is Empowerment a Protective Factor?Leela VisariaSocial history and novels in many Indian languages record violence against married women in India, mainly perpe-trated by their husbands. Yet, only in the last two decades has a systematic effort been made to estimate the magnitude of violence, its determinants and reasons, the forms inwhichit is manifested, and its health, social, legal and economic consequences. IntroductionSurvey-based studies have indicated that anywhere from 35 to 75% of women in India face verbal, physical, or sexual violence from their partners or other men known to them (see Jejeebhoy 1998; Mahajan 1990; Karlekar 1998; Jain et al 2004; Visaria 2000). Qualitative in-depth studies have thrown light on a range of issues such as women’s support-seeking behaviour, intergen-erational effects, the culture of silence, and the adherence to social norms that encourage tolerating, accepting, and even rationalising domestic violence for the sake of preserving family honour (Hassan 1995; Miller 1992; Jaisingh 1995; Koenig et al 2006). However, most of these studies were conducted with small samples and the findings could be generalised even for the states in which they were conducted. Also, very few studies have been carried out to examine these issues from the perspective of the perpetrators of violence. To overcome this limitation, the second National Family Health Survey (NFHS-2), conducted in 1998-99, took the bold step of nationally canvassing a few questions related to domestic violence, essentially to assess whether women in a large survey would respond to them.1 The questions were rather general and tried to measure the prevalence of violence and understand the situations in which ever-married women justified wife-beating. The respondents were read out six situations where wives trans-gressed their traditionally accepted roles or social norms. Women’s responses were sought on whether their husbands were justified in beating them if they deviated from their perceived “duties”.2 The success with canvassing these relatively sensitive questions emboldened the coordinators and advisers of the third NFHS (2005-06) to canvass an entire module on domestic violence with 25 key questions besides the ones on wife-beating (IIPS and Macro International 2007). InNFHS-3, two more situations were added and one dropped in the wife-beating question. The two new situations were arguing with the husband and refusing to have sexual intercourse with him. The situation of the woman’s family not providing the expected money, jewellery or other goods (implying dowry) was dropped This study examines three issues related to domestic violence in India on the basis of data from the third National Family Health Survey. It shows that a significant proportion of women, regardless of their socio-economic background, accept power differentials based on sex and that men have a right to discipline them. It also reveals that men from violent homes are significantly more likely to use violence against their wives. Most women do not seek help for the violence suffered, and the few who do, tend to steer clear of social service organisations and the authorities.Leela Visaria (lvisaria@gmail.com) is at the Gujarat Institute of Development Research, Ahmedabad.
NATIONAL FAMILY HEALTH SURVEY-3Economic & Political Weekly EPW november 29, 200861because less than 7% of the women inNFHS-2 cited it as a reason for violence. The same set of questions on attitudes towards wife-beating was asked to both women and men inNFHS-3. Interestingly, nearly the same proportion of women (54%) and men (51%) agreed that wife-beating was justified in at least one of the situa-tions. Also, the result was similar when the background charac-teristics of the respondents were considered. Among both men and women, a greater proportion of those who were illiterate, poor, young, and those in rural areas who were divorced or deserted justified wife-beating more than the others. It suggests that there is good agreement between men and women on the norms that govern married life and that if women fail to observe the norms they would bring violence upon themselves. The questions in the domestic violence module in the NFHS-3 sought information on the various forms of violence (physical, sexual or emotional) and the ways in which they were inflicted on all women, regardless of their marital status. Questions about the drinking habits of the partner (to understand its relationship to violence), spousal violence, help-seeking behaviour, the physi-cal consequences of spousal violence, and knowledge of parental violence were asked only to ever-married women. In addition, ever-married women were asked whether marital control was exercised in six specific situations. Some of the questions related to lifetime violence and some had a reference point of one-year. The responses to most of these questions were examined against the background characteristics of the women, and the data help understand some of the barriers to empowerment of women in the country. The results of this enquiry are presented in a separate chapter in the NFHS-3 report.This paper largely examines data from the NFHS-3. Rather than examining all the data collected on violence, I have examined findings on three issues which have not often been empirically studied in depth in India.3 The three issues are: degree of marital control exercised by husbands and other empowerment indicators of women and their relationship to violence – the association between knowledge of parental violence and experience and acceptance of spousal violence; and the extent to which women seek or expect to receive help to stop spousal violence. TheNFHS-2 reported that 21% of ever-married women experi-enced violence at some time since the age of 15 while the NFHS-3 says that 33.5% of all women aged 15 to 49 have done so.4 However, one cannot, on the basis of the increase in the preva-lence of violence between the two NFHS surveys, surmise that domestic violence has been increasing in India on the basis of these figures alone. The NFHS-2 included only a few questions to ascertain whether it was possible to collect data on violence in a large survey. There was also a strong likelihood of substantial under-reporting because women would have been very reluctant to respond positively to experiencing violence if the questions were asked in the presence of family members and others. The investigators were instructed not to ask the questions on violence, which were part of the ever-married women’s questionnaire, in the presence of others. However, it is difficult to know how far they succeeded in adhering to this instruction. In theNFHS-3, a separate module was prepared and canvassed only to one woman in each household, not to all the eligible women if there were more than one. More importantly, the respondents were clearly instructed to answer the violence module only if they were ensured complete privacy.5 Although estimates of lifetime physi-cal violence are available for all the states as well as the background characteristics of women respondents in the NFHS-2 andNFHS-3, it would not be prudent to discern a time trend given the variations in the mode of data collection. I shall, therefore, only examine the data available from the NFHS-3, the most recent survey.Aspects of Violence against WomenTo contextualise the three issues that are examined in the follow-ing sections, I briefly present some of the salient findings on the extent of violence reported by women and the differentials in accordance with their background characteristics and the region in which they live. Compared to women in the 15 to 24 age group, a greater proportion of older women experienced lifetime violence. A higher percentage of rural women (36%) were subjected to violence than women living in urban areas (28%), and a significantly greater proportion of women with little or no education experienced violence compared to their educated sisters. Only 14% of women with 12 or more years of schooling reported experiencing violence while the figure for illiterate women was 44%. Violence was reported by a higher percentage of currently married women (37.4%) than never married women (16.1%). But 66% of divorced, separated or deserted women reported having experienced physical violence. Violence was much more prevalent among women from the scheduled castes and tribes (39-42%) than among those from the higher castes (27%). Violence was also inversely related to the wealth index. Violence against women varied hugely among the states. More than 40% of women aged 15 to 49 reported having experienced physical or sexual violence in the relatively backward states of Rajasthan, Madhya Pradesh, and Bihar. Interestingly, among the big states, the percentage was only a tad lower in West Bengal, Uttar Pradesh and Tamil Nadu. On the other hand, less than 20% of women reported experiencing violence in states like Himachal Pradesh, Kerala, and Karnataka. It appears that rather than the economic development of a region, it is the attitude towards women, societal norms and perceptions about theirworthand position in the household, and men’s notions of self-esteem that influence husbands’ behaviour, for better or worse. In spite of these differentials, it is important to note that one in five women from the wealthiest group and one in seven women with educa-tion of 12 or more years reported being subjected to violence within the home, almost always by the spouse. A relatively high percentage of Tamil women reported violence compared to those from many other states and this merits atten-tion. An in-depth study undertaken in the slums of Tamil Nadu’s capital, Chennai, reported that the men held that women should be disciplined. They required their wives to be chaste, submissive, respectful and accepting of their imperfections (Go et al 2003). Wife-beating was condoned to ensure that women behave themselves and stay under the control of men. According to 
NATIONAL FAMILY HEALTH SURVEY-3november 29, 2008 EPW Economic & Political Weekly62NFHS-2, while 21% of ever-married women in the country as a whole said they had experienced lifetime violence, the percentage for Tamil Nadu was 40, the highest in the country (IIPS and ORC-Macro, Tamil Nadu, 2001). According to theNFHS-3, almost the same percentage (39%) of Tamil women reported experienc-ing physical or sexual violence. Further, acceptance of violence as a justifiable behaviour was also greater in Tamil Nadu. According to theNFHS-2, in the country as a whole, about 56% of ever-married women said their husbands were justified in beating them if they failed to perform certain duties such as showing respect to their in-laws or attend-ing to their children and household. In Tamil Nadu, 72% of women accepted violence as a justifiable act, thereby testifying that gender inequality was quite pervasive there (Kishor and Gupta 2004). According toNFHS-3, nearly two in three Tamil women agreed that in certain situations husbands were justified in beating their wives. An in-depth exploration of the acceptance of domestic violence by the women of Tamil Nadu would help understand the cultural and social norms that govern conjugal relationships in Tamil society. However, there are almost no such behavioural guidelines or restrictions apply to men in any Indian society. Risk Factors Husbands tend to exercise control over their wives by clearly in-dicating how they should behave. InNFHS-3, information was sought from ever-married women on six specific situations: whether the husband became jealous or angry if they talked to other men; whether the husband accused them of being unfaith-ful; whether the husband would not permit them to meet their female friends; whether the husband tried to limit contact with their natal family; whether the husband insisted on knowing where they are at all times; and whether the husband did not trust them with money. These situations reflected different di-mensions of women’s lives, ranging from economic independence and mobility to freedom to interact with friends and men known to them without arousing suspicion. While many women may not personally approve of such controlling behaviour, their accept-ance of it or inability to reject it indicate they are not empowered even within their marital home. The responses of women, classified by their background characteristics, are presented in Table 1. A little over a quarter of ever-married women reported that their husbands (or ex- husbands if the respondents were not currently married) became angry or jealous if they talked to other men. Some micro studies have also reported this controlling behaviour to be fairly widespread in many states of India (Jain et al 2004; Visaria 2000). Husbands show their anger or displeasure even when women reportedly talk to their brothers, male cousins or other male relatives from their natal villages, or neighbourhoods in the case of urban areas. A married woman having a male friend or visitor almost amounts to blasphemy and it becomes the subject of gossip not only in rural areas but also in many urban settings. A greater proportion of younger, recently married women in rural areas, with little or no education, who belonged to poor house-holds, and those who were divorced or separated, faced the wrath of their husbands (or ex-husbands) on this count than better educated older women or urban women. The logic seems to be that women’s behaviour has to be checked when they are young and newly married so that they learn to behave (that is, not be familiar or friendly with other men) according to social or family norms. The anger, jealousy or suspicion of husbands sometimes manifests itself in accusations that the wife is being unfaithful or having illicit relations with other men. Nearly 9% of women reported that they were frequently accused of being unfaithful. Again, the differentials were in the same direction as with jealousy. The husbands of nearly 12% of women insisted that they know where their wives were at all times. This desire to know their wives’ every movement largely stems from wanting to check that they do not see other men or talk about family problemsto others. Less educated younger women, those living in rural areas, and those divorced, separated or remarried from the lower wealth group faced this more than women belonging to other categories.6 Further, nearly 16% of women reported that they were not permitted to meet even female friends. In areas or communities where exogamy is practised and marriages are arranged by parents or other elders, women do not always have many acquaintances from their places of origin or areas near their marital home. The restriction enforced on women interacting with other women known to them is a very harsh controlling measure. Almost 10% of women also reported that their husbands tried to limit their contacts with members of their natal families. This is manifested by not allowing women to visit their natal family, except when it is absolutely essential, or not welcoming members of their family, or showing displeasure when they visit. Table 1: Percentage of Ever-Married Women Aged 15-49 Whose Husbands Exercised Marital Control, according to Select Background Characteristics( 2005-06) Percentage of Women Whose Husband Background Is Jealous Accuses Her Does Not Tries Insists on Does Displays Displays Characteristics If She Talks of Being Permit Her Limiting Knowing Not Trust Three or None of the to Other Men Unfaithful to Meet Her Contact Where Her with More Specific Female with Natal She Is All Any Specific Behaviours Friends Family the TimeMoney Behaviours Age 15-19 33.6 11.7 19.3 12.3 16.1 20.9 17.2 49.5 20-24 27.7 8.2 15.9 10.3 12.7 18.5 12.7 56.3 25-29 27.1 8.7 16.1 9.8 12.7 18.0 12.4 55.9 30-39 25.8 8.5 15.9 10.0 11.2 18.8 12.0 57.5 40-49 23.5 7.7 14.9 8.7 10.4 17.1 10.4 59.8Residence Urban 20.5 6.4 14.3 8.0 9.3 17.1 9.6 63.7 Rural 29.0 9.4 16.6 10.7 13.0 18.9 13.3 54.0Education Illiterate 32.3 1.08 16.5 11.4 13.4 19.3 14.5 51.5 < 5 years 28.3 9.7 19.7 11.6 14.8 21.2 14.8 53.112+ years9.6 2.6 12.4 4.5 6.2 14.05.0 73.8Caste/tribe SC 29.7 10.5 17.6 10.7 12.9 20.2 14.1 53.7 ST 31.3 12.0 17.6 11.0 15.3 18.9 16.0 52.7 OBC 27.2 8.2 13.6 9.8 11.1 16.1 11.5 58.3 Other 22.2 6.6 17.4 9.1 11.4 19.9 10.9 58.3Wealth index Lowest 33.9 12.417.212.815.419.816.049.3 Middle 28.8 10.0 16.0 10.5 12.3 17.9 13.0 55.6 Highest 14.8 3.0 14.0 5.8 7.1 16.6 6.8 68.1 All 26.4 8.5 15.9 9.9 11.9 18.3 12.2 56.9Not all educational or wealth index categories are presented in the table. Source: IIPS and Macro International 2007, Vol II, pp 505-06.
NATIONAL FAMILY HEALTH SURVEY-3Economic & Political Weekly EPW november 29, 200863When women are married to men from the same village or town, they may experience greater freedom or find ways to visit or interact with their family members without being noticed. But when distance is a factor, this restricting or controlling behaviour tends to have a detrimental effect on women, mainly because they have no opportunities to share their problems with their family members or near ones. Not granting even a modicum of economic independence, especially to women who have no other source of income, is another controlling behaviour. It leads to husbands saying women cannot be trusted with money, implying that they do not know how to spend money judiciously. A little over 18% of ever-married women indicated that their husbands do not trust them with money. This controlling behaviour is expressed by asking women to explain how every rupee is spent and reprimanding them if something is spent on what husbands consider unnecessary. Background characteristics hardly made a difference in the case of this controlling behaviour, which stems from the general notion that women are not careful about what they spend money on. The NFHS-3 estimates that 12% of women reported three or more controlling behaviours by their husbands. The differences when background characteristics were taken into account were not very significant, except that women from poor households faced more controlling behaviour than those belonging to better-off households. Instead of grouping any three types of controlling behaviour, it would be interesting to group the three types of behaviour that prompt husbands to suspect and mistrust their wives when they deal with other men, even their male kin. In a sense, such behaviour undermines the very basis of a marital relationship. Interestingly, the differences between women belonging to various socio-economic groups were very small when it came to restrictions on meeting female friends and handling money. The former controlling behaviour very likely stems from the fear that women will share news about family matters that husbands or in-laws do not want divulged to outsiders. The underlying fear is that women may do so until their loyalty to the family of their in-laws is established. So, young women, even educated ones, are not trusted. Table 1 also shows that 57% of women reported that their husbands do not display any of such specific controlling behav-iour, implying that they are trusted by their spouses. Conversely, 43% of women reported that their husbands show at least one type of controlling behaviour, and they were asked their opinion about this. As expected, the extent of trust was greater among older women (some of whom may have experienced greater control over their behaviour when they were younger; with the passing of time, they gain the trust of their husbands and in-laws), better educated women and among those belonging to better-off households. The extent of violence experienced by women was also examined on the basis of some characteristics of the husband and selected indicators of women’s empowerment. The data presented in Table 2 show that a greater proportion of husbands who are either illiterate or have little education inflict violence (physical, sexual or emotional) on their wives than husbands who are better educated. All the same, women’s education has a much stronger association with violence than men’s education. One in four men with 12 or more years of schooling used violence against their wives but only 15% of women with the same level of education reported being subject to violence by their spouses. It is only education beyond 12 years of schooling that appears to empower women and act as a protective factor. Also, as evident in Table 2, women who have the same level of education as their husbands are least likely to suffer either physical or sexual violence compared to those who are illiterate or have less education than their husbands. Drinking alcohol is very significantly associated with both physical and sexual violence. Seven out of 10 men who get drunk subjected their wives to violence as against three out of 10 men who did not drink. Also, a quarter of men who get drunk inflicted sexual violence on their wives. The combination of sexual desire and alcohol increased women’s risk to violence if she refused sex. A study conducted in south India indicated that the risk of wife abuse increases significantly with alcohol consumption by the husband (Rao 1997). Another study carried out in Karnataka reported that independent of caste and economic status, alcohol consumption by husbands was found to be significantly associa-ted with violence (Krishnan 2005). More than moderate consump-tion of alcohol by men definitely increases the chances of violence against women. In addition, 81% of men who displayed five or all of the six controlling behaviours inflicted physical or sexual violence on their wives. Two out of five men who exercised a high degree of marital control also sexually violated their spouses. The Table 2: Percentage of Ever-Married Women Aged 15-49 Whose Husbands Have Subjected Them to Physical or Sexual Violence, according to Select Husbands’ Characteristics, and Empowerment Indicators(2005-06)Characteristics PhysicalViolenceSexual Violence Physical or Sexual ViolenceHusband’s education Illiterate 45.2 12.8 47.2 < 5 years 43.8 13.0 45.7 8-9years 34.2 10.4 36.6 12 + years 18.9 4.9 20.8Husband’s alcohol consumption Does not drink 28.0 7.9 30.3 Gets drunk sometimes 46.8 11.6 48.5 Gets drunk very often 67.6 23.6 68.6 Spousal education difference Husband better educated 34.1 9.7 36.4 Wife better educated 29.7 9.0 31.9 Both equally educated 21.4 6.0 23.3 Bothilliterate 46.0 12.8 47.8Number of marital control behaviours displayed by husband None 23.9 5.0 25.6 1-2 44.1 12.7 46.8 3-4 60.2 23.2 63.0 5-6 77.6 39.1 80.9Number of decisions in which women participate None 32.9 11.6 36.1 1-2 35.9 11.0 38.3 3-4 34.4 8.4 36.1Number of reasons for which wife-beating is justified None 28.2 8.4 30.4 3-4 42.0 11.6 47.7 All7 41.4 11.9 42.8 All ever-married women 35.1 10.0 37.2Source: IIPS and Macro International 2007, Vol II, pp 511-12. 
NATIONAL FAMILY HEALTH SURVEY-3november 29, 2008 EPW Economic & Political Weekly64controlling behaviours stem from a lack of trust in women and leads to violence against them.To understand women’s participation in decision-making in the household, the NFHS-3 asked women whether they participatedin decisions related to their own health, major household purchases, purchases for daily household needs, and on visits to their family and relatives. If women did not participate in any of these decisions, they received a score of zero. Those who participated in one or two decisions were viewed as moderately empowered and those who had a voice in three or all four decisions were viewed as highly empowered. As evident in Table 2, empowerment has no clear relationship to prevalenceof violence. This belies the expectation that women who participate in house-hold decisions, and therefore have egalitarian gender-role attitudes, are less likely to experience violence. On the other hand, 42-44% of women who indicated that wife beating was justified in any one of the six situations were themselves recipients of physical or sexual violence compared to 30% of those who said none of the situations excused violence. Overall, well-educated women and women in marital relationships where the husbands did not display controlling behaviours were the ones most likely to escape violence. Intergenerational EffectIn theNFHS-3, ever-married women were asked a question on whether their mothers were beaten by their fathers. Responses to this reflected the extent to which young girls who witness parental violence, knowingly or unknowingly, accept violence as a part of their own married life. The data presented in Table 3 show that two-thirds of women who knew that their mothers were beaten by their fathers experienced some form of violence at the hands of their husbands. For nearly 60%, the violence was physical or sexual. The probability of children who have witnessed parental violence inflicting the same on their spouses when they grow up is quite high and a cause for concern. A third of women who said they had not seen any parental violence also reported being subjected to violence by their husbands. It is likely that some women who said that they were not aware of, or did not know about, parental violence were reluctant to disclose what happened between their parents. A fairly large study carried out to understand the behaviour of men in Uttar Pradesh reported that husbands who had witnessed their fathers beating their mothers as children were 4.7 times more likely to beat their own wives and three times more likely to sexually coerce them than men who had not witnessed such violence (Koenig et al 2006). Martin et al (2002) showed that witnessing violence between one’s parents while growing up is an important risk factor for the perpetra-tion of violence on one’s partner in adulthood. Compared to men raised in non-violent homes, men from violent homes were significantly more likely to believe in the right to control their wives and to physically and sexually abuse them. The study also demonstrated that non-violence in the earlier generation was strongly predictive of non-violence in the second generation.Although theNFHS-3 included only one question on the inter-generational effect of domestic violence, this is an area which needs much more exploration in-depth. The effect of witnessing violence on the minds of children, the internalisation of the prevailing norms related to violence, the subsequent behaviour and the rationalisation of that behaviour all need to be investi-gated while addressing the issue of violence and ways to break the cycle of violence. Help-Seeking Behaviour In the NFHS-3, all women who reported physical or sexual violence were asked a number of questions on whether they sought help to try and end the violence. Women who said that they did seek help were asked from whom they sought it. Also, those who reported that they did not seek any help were also asked questions on whether they took anyone into confidence and shared their plight with them. Table 4 shows some of the data classified according to the background characteristics of the women. Only about one in four women (23.8%) sought help to end the violence they experi-enced. Two out of three women neither sought help nor didtheytellanyone (family members or friends) about experi-encing violence. What is quite striking is that there are virtually no differentials in telling others about the violence, or seeking help from someone, whatever the background characteristics. Neither education nor family wealth act as protective factors in this Table 3: Percentage of Ever-Married Women Aged 15-49, Who Reported that Their Father Beat Their Mother, Experiencing Violence by Their Husbands(2005-06)Respondent’s Father Emotional Physical Sexual Any of the Beat Her Mother Violence Violence Violence Three FormsYes 28.5 57.4 59.6 62.2No 12.4 28.3 30.3 32.8Do not know 18.0 46.5 49.7 52.1Source: IIPS and Macro International 2007, Vol II, p 510.Table 4: Per Cent Distribution of Women Aged 15-49 Years Who Told Someone about Experiencing Violence and Sought Help to End It, according to Background Characteristics(2005-06)Background Characteristics Never Told Anyone % Who Told Someone Sought Help from Any SourceResidence Urban 65.2 8.6 23.6 Rural 66.4 7.3 23.9Education Illiterate 66.5 7.8 23.7 < 5 Years 65.9 7.8 24.0 8-9 Years 66.9 8.2 22.3 12 or more years 69.0 6.8 20.0Caste/tribe SC 64.8 7.0 26.1 ST 65.4 9.5 23.2 OBC 65.0 7.8 24.6 Other 69.2 7.3 20.9Wealth index Lowest 67.1 8.0 23.2 Middle 66.1 7.1 24.8 Highest 67.8 8.1 20.4Type of violence Physicalonly 68.1 7.5 21.7 Sexualonly 85.3 4.1 7.7 Both physical and sexual 52.7 9.5 36.7 All 66.1 7.7 23.8Source: IIPS and Macro International 2007, Vol II, pp 522-23. 
NATIONAL FAMILY HEALTH SURVEY-3Economic & Political Weekly EPW november 29, 200865regard. In fact, better educated women and those belonging to families with a better economic status were more unlikely to share their experience of violence with others. Women who experienced sexual violence were even more reticent about talking to others about it or seeking help from anyone. The silence surrounding violence in general and sexual violence in particular has been reported in other micro and in-depth studies (Visaria 2000 and Visaria 2002). This has to be understood in the context of women trying to preserve the family’s honour by not disclosing something that happened within the home, and also in the context of the sense of shame associated with being abused by someone known to them and with whom they share intimate or conjugal relations. In such a situation, from whom do battered women seek support? In theNFHS-3, ever-married women were asked this question. The majority of women who experienced violence and sought help reported that they did so from their natal family; 71% turned to their parents and other family members for support. Nearly 30% sought help from husbands’ families.7 Neighbours were resorted to by 15% and 9% turned to friends. Many a time, neighbours witness the violence and occasionally also get involved in an attempt to ease the situation. Hardly any women opted to report cases of violence to formal organisations or the authorities, presumably because they feared being ostracised and shamed by the communities in which they live. The fear that they themselves will be blamed for provoking husbands to use violence is all too real. It is the reality of Indian society that women who summon the courage to challenge their abusers in a court of law or seek the support of social service organisations have to be prepared to face a long and humiliating battle with little sympa-thy from the authorities or family members, and even the media. Interviewing women survivors of abuse in New Delhi, Prasad (1999) demonstrated that the legal system and procedures designed to enhance women’s access to the law actually inhibited it, and that the state as such showed a tolerance of domestic and sexual violence.ConclusionsIn this paper, I have largely examined three issues related to domestic violence in India. They are the degree of marital control exercised by husbands, and other empowerment indicators of women and their relationship to violence, the association between knowledge of parental violence and experience and acceptance of spousal violence, and the extent to which women seek or expect to receive help to stop spousal violence. The underlying question is whether women’s empowerment is a protective factor that safeguards them against violence. The findings show that a significant proportion of women, regardless of socio-economic background, subscribe to power differentials based on sex and accept that men have a right to discipline them, especially when they fail to fulfil gender- specific duties such as taking care of the house and children or cooking food on time in a manner that pleases the husband. Further, women who are beaten or otherwise physically abused tend to justify their husbands’ behaviour, as a way of rational-isingthe treatment meted out to them.Admittedly, the subordination of women and the dominance of husbands when they are perceived to be transgressing from their wifely duties is not unique to India and cuts across cultures and nations. Nonetheless, the experience of violence, or even the threat of violence, and controlling behaviour by their spouses tends to lower women’s self-esteem, instil fear in them and further lower their ability to fulfil daily tasks to the satisfaction of the members of the family. The controlling behaviour that prompts husbands to suspect the moral character of their wives and mistrust their dealings with other men, including their male kin, undermines the very basis on which a marital relationship rests.Witnessing violence between one’s parents while growing up has been found to be an important risk factor for the perpetra-tion of violence on partners in adulthood. Men from violent homes are significantly more likely to believe in husbands’ rights to control their wives, and to be physically and sexually abusive towards them. The internalisation of the prevailing norms related to violence, and the subsequent behaviour and rationalisation of that behaviour need to be examined while addressing the issue of violence and the means to break the cycle of violence.As in other surveys, in the NFHS-3 hardly any women reported that they sought redress or support from formal organisations or the authorities to deal with the violence meted out to them for fear of being ostracised and shamed by the communities in whichthey live. The fear that they themselves will be blamed for provoking men to use violence against them looms large. In the absence of supportive shelters or other avenues, it is very difficult for battered women in India to gather the courage to challenge their abusers in a court of law or seek the support of the few social service organisations that exist. The humiliation that those who approach the judiciary face because of long court battles, and having to deal with an insensitive police and others with little sympathy unnerve most women. They choose to suffer in silence in their homes, which become dysfunctional.Even education does not empower women to enter the public arena for support. Better educated women or those belonging to better-off families who experience violence are least likely to share their experiences or seek support from others. This needs to be understood in the context of a culture of silence where women try to not disclose what happens within the home environment. Equally important is the sense of shame associ-ated with being abused by someone known to them and with whom they share intimate or conjugal relations. Even when physically injured, women remain silent and suffer it alone. Further, societal norms that tolerate and accept violence are widely prevalent in Indian society and an adherence to them prevents women from seeking care. Finally, an in-depth exami-nation of the widespread acceptance of spousal violence among women of Tamil Nadu would help understand the prevalent cultural and social norms that govern conjugal relationships in Tamil society.
NATIONAL FAMILY HEALTH SURVEY-3november 29, 2008 EPW Economic & Political Weekly66Notes 1 The question in the NFHS-2 which asked whether a woman had been beaten or physically mistreated ever did not offer any definition of what consti-tuted violence. So a woman for whom an occasional slap from her husband qualified as routine or normal behaviour was not likely to regard it as an act of physical violence and report it as such.2 The situations in the NFHS-2 wife-beating question were: the husband suspecting his wife of being unfaithful, the wife’s family not giving the expected amount, jewellery or other goods, the wife showing disrespect to her in-laws, the wife going out without telling the husband, the wife neglectingthehouseor children, and the wife not cooking food properly (IIPS and ORC Macro 2000, p 72). 3 The prevalence of life-time physical violence against Indian women has been estimated by a number of research studies. Similarly, the forms of violence and the reasons for them have also been explored at some length. Some relatively large studies have also examined these against the backgrounds of the respondents to under-stand whether there are differentials in violence experienced by women by place of residence, age, education, employment status or social group.4 Unlike the NFHS-2, which collected information from ever-married women, the NFHS-3 asked some of the violence-related questions to all women, including single women aged 15 to 49 years. 5 For the first time, in the NFHS-3, men were also interviewed and questions were asked on whether they were justified in beating their wives in certain situations, and whether they had the right to get angry or reprimand their wives if they refused sex. These data are available in the NFHS-3 report but are not analysedinthispaper.6 One may argue that one of the reasons for divorce or separation is that women behave freely with other men and thereby invite the wrath or suspi-cion of their husbands. One can equally argue that it is the controlling and suspicious nature of husbands that stifle women’s freedom which results in divorce/separation. A very sensitively conducted in-depth study would help unravel such complexities in spousal relationships. 7 Women reported seeking help from multiple sources and so the figure can exceed 100.ReferencesGo, V F et al (2003): “When HIV-Prevention Messages and Gender Norms Clash: The Impact of Domestic Violence on Women’s HIV Risk in Slums of Chennai, India”, Aids and Behaviour, Vol 7, No 3, pp 263-72.Hassan, Yasmeen (1995):The Haven Becomes Hell – A Study of Domestic Violence in Pakistan, Special Bulletin (Lahore: Shirkat Gah). IIPS and Macro International (2000):National Family Health Survey (NFHS-2), 1998-99: India (Mumbai: International Institute for Population Sciences). – (2001): National Family Health Survey (NFHS-2), India, 1998-99: Tamil Nadu (Mumbai: Inter-national Institute for Population Sciences). – (2007): National Family Health Survey (NFHS-3), 2005-06, India,Vol I (Mumbai: International Institute for Population Sciences). – (2007): National Family Health Survey (NFHS-3), 2005-06, India, Vol II (Mumbai: International Institute for Population Sciences).Jaisingh, I (1995): “Violence against Women: The Indian Perspective” in J Peters and A Wolper (eds), Women’s Rights, Human Rights (New York: Routledge).Jejeebhoy, Shireen (1998): “Wife Beating in Rural India: A Husband’s Right? Evidence from Survey Data”,Economic & Political Weekly, Vol 33, No 15, 11 April, pp 855-62.Karlekar, M (1998): “Domestic Violence”,Economic & Political Weekly, Vol 33, No 27. Kishor, S and K Gupta (2004): “Women’s Empower-ment in India and Its States: Evidence from the NFHS”,Economic & Political Weekly, Vol 39, No 6, pp 694-712. Koenig, M A et al (2006): “Individual and Contextual Determinants of Domestic Violence in North India”,American Journal of Public Health, Vol 96, pp 132-38.Krishnan, S (2005): “Gender, Caste, and Economic Inequalities and Marital Violence in Rural South India”, Health Care for Women International, Vol 26, No 1, pp 87-99. Mahajan, A (1990): “Instigators of Wife Battering” in Sushama Sood (ed), Violence against Women (Jaipur: Arihant Publishers), pp 1-10.Martin, S L et al (2002): “Domestic Violence across Generations: Findings from Northern India”, International Journal of Epidemiology, Vol 31, No 3, pp 560-72. Miller, Barbara D (1992): “Wife-beating in India: Variations on a Theme” in D A Counts, J K Brown and J C Campbell (eds),Sanctions and Sanctuary: Cultural Perspectives on the Beating of Wives (Boulder: Westview Press), pp 173-84.Prasad, S (1999): “Medicolegal Response to Violence against Women in India”, Violence against Women, Vol 5, No 5, pp 478-506.Rao, Vijayendra (1997): “Wife-beating in Rural South India: A Qualitative and Econometric Analysis”, Social Science and Medicine, Vol 44, No 8, pp 1169-80.Visaria, L (2000): “Violence against Women: A Field Study”, Economic & Political Weekly, Vol 35, No 20, May 13, pp 1742–51. – (2002): “Many a Slip between Slap and Lip”, The Little Magazine, Vol 3, Issue 4, pp 40-43.Hind Swaraj Centenary International SeminarSocial Development and Human Civilisation in the 21st Century12-14 February 2009, India International Centre, New DelhiJointly organised by Council for Social Development, India International Centre, DCRC, Delhi University, CPS, Jawaharlal Nehru University, Nelson Mandela Centre, Jamia Milia Islamia, ISHS, University of South Africa, REGGEN, Federal Flomenance University of Brazil & Global Studies, University of California, Santa BarbaraCall for PapersSeveral institutions have come together to organise a major international seminar on the occasion of the centenary of Gandhi’s Hind Swaraj. The objective is to take Hind Swaraj as a vantage point and engage ourselves on some of the fundamental questions our age and try to make a contribution to basic thinking on those issues. The following panels have been proposed to reflect on challenges of the 21st century from the perspective of people of Asia, Africa and Latin America. • Development Theory •Peace • World Order • Civilization and Culture • Self Determination, Liberation and Freedom • Science, Technology and Swaraj • Human Rights and Social Justice • Religion and Social Imagination • Feminism • Swaraj, State and the Marginalised • Field reports on Gram Swaraj experiments • Hind Swaraj as a classical text in world literature Abstracts of 250 words are invited from scholars and activists for the above panels by 15 December 2008. Authors of selected papers will be intimated by 31 December 2008. The last date for the submission of full length papers of 5000 to 6000 words is 31 January 2009. Local hospitality will be provided to all selected participants. Limited travel support is available for international participants and for scholars and activists within India. Contact: Seminar Convener, Council for Social Development53 Lodhi Estate, New Delhi-110003, IndiaPhone: 91-11-24615383, 24611700, 24616061 (Fax) Emails: swaraj@csdindia.org, dr_mohanty@yahoo.com, csdnd@del2.vsnl.net.in
