|Violence against Women (World Bank, 1994, 84 pages)|
Gender violence has important implications for socioeconomic development and for key initiatives already high on the international health agenda. Yet few mainstream development organizations, even among those devoted to health issues, have focused on violence. The World Health Organization and a handful of NGOS have supported efforts to eliminate genital mutilation, but such abuses as battery, rape, and incest have been largely ignored. The Women, Health and Development program of the Pan American Health Organization (PAHO)) and the Canadian NGO MATCH International are notable exceptions
Effect on socioeconomic development
Gender violence, through its effects on a woman's ability to act in the world, can serve as a brake on socioeconomic development. The development community has come to realize that such problems as high fertility, deforestation, and hunger cannot be solved without women's full participation. Yet women cannot lend their labor or creative ideas fully when they are burdened with the physical and psychological scars of abuse.
New evidence from the United States suggests that the scars of victimization can also lead to lower future educational attainment and income levels far women who are abused. Using simultaneous equations to model the income effects of childhood sexual abuse, Batya Hyman (1993) has shown that women who have been abused earn 3 to 20 percent less each year than women who have not been abused, with the discrepancy depending on the type of sexual abuse experienced and the number of perpetrators (the model controls for all other factors known to affect income prospects). Incestuous abuse affected income indirectly through its impact on educational attainment and mental and physical health status. Women sexually abused by strangers suffered an additional direct effect on income; Hyman speculates that they learn from this abuse that the outside world is dangerous, and therefore limit their engagement in the world.
Violence against women can also thwart the development of the wider community through its effect on women's participation in development projects. A study commissioned by UNIFEM/Mexico to find out why women stopped participating in projects found that threats from men were a major cause. Men perceived the growing empowerment of their wives as a threat to their control, end used beatings to try to reverse this process of empowerment. In Madras, India, a revolving loan fund of the Working Women's Forum almost collapsed after the project leaders, subjected to increased domestic violence, stopped participating (Carrillo 1992). As Dr. Christine Bradley, Principal Project Officer for the Papua New Guinea Law Reform Commission, observes:
Simply attending a meeting may be dangerous for a woman whose husband does not want her to go in [Papua New Guinea] some husbands prevent their wives from attending meetings by locking them in the house, or by pulling them off the vehicle they have boarded to take them to the meeting, or even by pursuing them to the meeting and dragging them home. (Bradley 1990, p. 5)
In a particularly gruesome example of male backlash, a female leader of the highly successfully government sponsored Women's Development Programme in Rajasthan, India, was recently gang raped by male community members because they disapproved of her organizing efforts against child marriage. They raped the woman in her home in front of her husband, and warned him Keep your wife in line or we'll rape her again " The incident and the fear that it induced dealt a major blow to the project's momentum (Rao Gupta, personal communication, 1993; Mathur 1992).
Elsewhere, men may use force to divert the benefits of development from women. Case studies of victims of domestic violence in Peru and of garment workers in the Mexican maquiladoras reported that men frequently beat their wives to get their earnings (Vasquez and Tamayo 1989, as quoted in Carrillo 1992).
To avoid violence, women learn to restrict their behavior to what they think will be acceptable to their husbands or partners. As Bradley (199O) observes, "Threats or fears of violence control women's minds as much as do acts of violence, making women their own jailers" In Papua New Guinea, for example, a recent study reports that married female teachers do not apply for or accept promotions in large part because they fear retaliation from their husbands: women represent only 39 percent of the country's primary school teachers and 5 percent of head teachers (Gibson 1990).
Fear of stranger-perpetrated violence similarly limits women's participation in public life. In the United States 49 percent of 299 women surveyed in six neighborhoods in Chicago, Philadelphia, and San Francisco estimated their chances of being raped in their own neighborhood as five or higher on a 10-point scale. Nearly half said that they relied on restrictive, isolating tactics (not going out, not going to certain places) "all or most of the time" or "fairly often" to protect themselves. By contrast, 90 percent of men living in the same neighborhoods said that they never restricted their behavior out of fear (Gordon and Riger 1989). Similarly, in a 1990 newspaper survey in Seoul, Korea, women identified fear of sexual violence as a principal cause of stress in their lives (Korea Sexual Violence Relief Center 1991). In a separate survey of 2,270 Korean women, 94 percent said that they felt uneasy because of the spread of sexual violence against women Forty percent felt "extremely uneasy" and reported restricting their activities because of their fear (Korea Sexual Violence Relief Center 1991).
In the developing world this distinctly female fear can have unexpected and insidious effects. Fear of rape has exacerbated underrnutrition among Ethiopian refugee families in Sudanese border camps. In a recent survey of women's mental health sponsored by the United Nations Development Programme, Ethiopian refugee women said that they had reduced the number of cooked meals they fed their children because they feared being raped-as many had been-while collecting firewood, a task requiring a two-to-three-hour foray outside the camp (LaPin 1992). Similarly, female health promoters working in rural Gujurat, India, when discussing obstacles to their work, emphasized their reluctance to travel alone between villages for fear of being raped. They requested self defense training to enable them to continue their work (Khanna 1992). These examples, far from isolated, illustrate the paralyzing and largely unrecognized effect that violence can have on women and on social development.
Effect on maternal health
Pregnancy should be a time when the health and wellbeing of women are especially respected. But surveys suggest that pregnant women are prime targets for abuse. Results from a large prospective study of battery during pregnancy among low-income women in Baltimore and Houston indicated that one of every six pregnant women was battered during her present pregnancy (McFarlane and others 1992). The study, published in the Journal of the American Medical Association, followed a stratified cohort of 691 white, African-American, and Hispanic women for three years. Sixty percent of the abused women reported two or more episodes of violence, and they were three times as likely as nonabused women to begin prenatal care in the third trimester. Other studies indicate that, compared with women who are not beaten, women battered during pregnancy run twice the risk of miscarriage and four times the risk of having a low-birth-weigh baby (Stark and others 1981; Bullock and McFarlane 1989). Low birth weight is a powerful predictor of a child's survival prospects in the first year of life.
In the developing world a survey of 342 randomly sampled women near Mexico City revealed that 20 percent of those battered reported blows to the stomach during pregnancy (Shrader Cox and Valdez Santiago 1992). And in a study of 80 battered women who sought judicial intervention against their partners in SanJose, Costa Rica, 49 percent reported being beaten during pregnancy. Of these, 7.5 percent reported miscarriages due to the abuse (Ugalde 1988). For mothers in developing countries who are already malnourished and overworked and lack access to adequate health care, battering during pregnancy is likely to have an even greater effect than for most women in industrial countries.
Violence may also be responsible for a sizable but unrecognized share of maternal mortality, especially among young unwed pregnant women. Fauveau and Blanchet (1989) report that in Matlab Thana, Bangladesh, homicide, and suicide motivated by the stigma of rape, pregnancy outside of marriage, or beatings or by dowry problems, accounted for 6 percent of 1,139 maternal deaths between 1976 and 1986. The figure rises to 21.5 percent if deaths due to botched abortions are included, many of which arc like wise related to shame over pregnancies outside of marriage. Among all deaths of women age 15 to 44 (not just maternal deaths), intentional injury accounts for 12.3 percent, with deaths due to homicide and suicide outnumbering those due to abortions.
Intentional injury was also found to be a significant cause of maternal deaths among women in Chicago, Illinois. Researchers examining the records of the Cook County Medical Examiner found that trauma was the number one cause of maternal deaths between January 1986 and December 1989, accounting for 46 percent of all maternal deaths. Of these, percent were duo to homicide and 9 percent to suicide (Fildes, Reed, and Jones 1992).
A prospective study of 161 women living in Santiago, Chile, revealed that women living in socially and politically violent areas had a significantly higher risk of pregnancy complications than women in less violent neighborhoods. Alter adjusting for potentially confounding variables (income, education, marital status, underweight, cigarette smoking, dissatisfaction with neighborhood, life events, alienation, uncertainty, and depression), researchers found that high levels of sociopolitical violence were associated with an approximately fivefold increase in risk of such pregnancy complications as preclampia, premature labor, threat of miscarriage, and gestational hypertension (Zapata and others 1992). If the stress and trauma of living in a violent neighborhood can induce complications, it is reasonable to assume that living in the private hell of an abusive relationship can produce similar sequelae.
Effect on family planning
Many women limit their use of contraception out of fear of male reprisal (Dixon-Mueller 1992). Men in many cultures reject birth control because they think it signals a woman's intention to be unfaithful, based on the logic that protection against pregnancy allows a woman to be promiscuous. And where fathering children is a sign of virility, a woman's request to use birth control may be interpreted as an affront to her partner's masculinity. Although the male partner's approval is not always the deciding factor, studies from countries as diverse as Bangladesh, Mexico, and South Africa have found that it was the single greatest predictor of women's contraceptive use. When partners disapprove, women forgo contraception or resort to family planning methods they can use without their partner's knowledge.
The unspoken reality behind this subterfuge is that women can be abused if they do not comply with men's sexual and childbearing demands. In a recent interview Hope Mwesigye of FIDA-Uganda, a nonprofit legal aid organization for women in Kampala, recounted the story of a young married mother running from a husband who beat her regularly. Although he earned a decent wage, the woman's husband refused to maintain her and their two children. To avoid bringing into the world more children whom she could not feed, the woman began using birth control, without her husband's consent. When she failed to bring forth more children, the beatings began; they became more brutal when he learned that she was using contraceptives (Banwell 1990).
Where legal provisions require spousal permission before birth control can be dispensed, women can be at increased risk of violence. According to Pamela Onyango of Family Planning International Assistance, women in Kenya have been known to forge their partner's signature rather than risk violence or abandonment by requesting his permission to use family planning services (Banwell 1990). Researchers conducting focus groups on sexuality in Mexico and Peru found that women there similarly feared violence, desertion, or accusations of infidelity if they brought up birth control (Folch-Lyon, Macorra, and Schearer 1981; Fort 1989). Not surprisingly, when family planning clinics in Ethiopia stopped requiring spousal consent, use of the clinics rose 26 percent in just a few months (Cook and Maine 1987).
Not all women who fear violence because of using or discussing contraceptives are necessarily at risk of actual abuse. In fact, some studies suggest that many men may be more accepting of family planning than most women suspect (Gallan 1986). But communication in marriage is often so limited that women have no idea of their partner's view of family planning but assume that it mirrors the cultural norm-frequently that men want large families and distrust women who use birth control. The discrepancy between women's perceptions and reality also speaks to the way that violence induces fear by example.
Even in countries where birth control is generally accepted, violence can restrict a women's ability to exercise reproductive and sexual autonomy. In a representative survey of women in Texas, more than 12 percent of the 1,539 respondents reported having been sexually abused by a husband, ex-husband, boyfriend, or ex-boyfriend after the age of 18. Of those 187 women, 12.3 percent stated that they had been prevented from using birth control and 10.7 percent that they had been forced to get pregnant against their will (Grant, Preda, and Martin 1989).
Studies from the United States suggest that sexual victimization may play an indirect role in perpetuating unwanted pregnancy. In a community based, random survey of women in Los Angeles, Wyatt, Gutherie, and Notgrass (1992) found that women who were sexually abused in childhood were 2.4 times more likely to be sexually revictimized during adulthood; revictimized women, in turn, had a significantly higher rate of unintended and aborted pregnancies than non-revictimized women.
Boyer and Fine's (1992) study of adolescent mothers in Washington State, discussed above in the section on the health effects of child and adolescent sexual abuse, suggests that there an: links between childhood sexual abuse and unwanted pregnancy among teenage women. Noting that concerted effort to improve teenagers' access to contraception and sex education had failed to reduce the rate of adolescent pregnancy in the United States over the past 20 years, the authors suggest that a "key factor in the conundrum of adolescent high-risk sexual behavior and adolescent pregnancy" (1992, p. 11) may be unresolved issues around early sexual victimization.
Effect on std and aids prevention
Not surprisingly, male violence can impede women's ability to protect themselves from HIV and other sexually transmitted diseases (STDs). Violence can increase a woman's risk through nonconsensual sex or by limiting her willingness or ability to get her partner to use a condom. In many cultures suggesting condom use is even more threatening than raising birth control in general, because condoms are widely associated with promiscuity, prostitution, and disease. A woman's act of bringing up condom use can be perceived as insinuating her infidelity or implicitly challenging a male partner's right to conduct outside relationships. Either way, it may trigger a violent response (Worth 1989).
An AIDS prevention strategy based solely on "negotiating." condom use assumes an equity of power between men and women that simply does not exist in many relationships. Even in consensual unions, women often lack control over their sexual lives. A study of home-based industrial workers in Mexico, for example, found that wives' bargaining power in marriage was lowest with regard to decisions about whether and when to have sexual intercourse (Beneria and Roldan 1987). Studies of natural family planning in the Philippines, Peru, and Sri Lanka and of sexual attitudes among women in Guatemala report forced sex in marriage, especially when men arrive home drunk (Liskin 1981; Lundgren and others 1992_. The summary of the Guntemalan study's focus groups observes that "it is clear from the replies the women gave...that being forced through violence to have sex by their partner is not an uncommon experience for Guatemalan women" (Lundgren and others 1992, p. 34).
For women who live with violent or alcoholic partners the possibility of coercive sex is even more pronounced. In the United States 10 to 14 percent of married women report being physically forced to have sex against their will, but among battered women the prevalence of coercive intercourse is at least 40 percent (Campbell and Alford 1989). In Bolivia and Puerto Rico 58 percent of battered wives report being sexually assaulted by their partner (ISIS International 1988), and in Colombia the reported rate is 46 percent (PROFAMILIA 1992). Given the percentage of women around the world who live with physically abusive partners, sexual coercion within consensual unions is probably common.
Childhood sexual abuse also puts individuals at increased risk of STDs, including AIDS, through the responses it generates in victims. Several studies link a history of sexual abuse with a high risk of entering prostitution (Finkelhor 1987; James and Meyerding 1777). Researchers from Brown University found that men and women who had been raped or forced to have sex during their childhood or adolescence were four times more likely than non-abused people to have worked in prostitution (Zierler and others 1991). They were also twice as likely to have multiple partners in any single year and to engage in casual sex with partners they did not know. Women survivors of childhood sexual assault were twice as likely to be heavy consumers of alcohol and nearly three times more likely to become pregnant before age 18. These behaviors did not translate directly into higher rates of HIV among women, but men who had experienced childhood sexual abuse were twice as likely to be HIV-positive as men who had not. The higher prevalence of HIV among male survivors could not be explained by a history of intravenous drug use.
Based on a probability survey of 407 men and women on Barbados, anthropologist Penn Handwerker has likewise shown that sexual abuse is the single most important determinant of high-risk sexual activity among Barbadian adolescents (Handwerker 1993a). After a wide range of socioeconomic and home-environment variables (for example, an absent father) are controlled for, sexual abuse remains strongly linked both to the number of partners adolescents have and to their age at first intercourse. Further analysis shows that the direct effects of childhood sexual abuse on a person's sexual behavior remain significant into the mid-thirties. For men, physical, emotional, and sexual abuse in childhood is also highly correlated with failure to use condoms in adulthood, after controlling for many other variables.
There is some evidence that sexual abuse may affect women's risk of AIDS through its effect on their drug use (Fullilove, Lown, and Fullilove 1992; Paone and Chavkin 1993). In a qualitative study of women attending an outpatient methadone maintenance clinic in the South Bronx, early sexual abuse-especially incest-emerged as one of the most formative experiences in the lives of women addicted to drugs (crack, cocaine, heroin). As author Dooley Worth explains:
The sense of stigmatization and shame experienced by female incest victims... leaves the young women feeling unloved, unlovable and unable to say "no" to things they do not want to do such as having sex or using drugs. (Worth 1991; emphasis in original)
Another effect of incest in girls is a tendency to dissociate from their bodies. This makes denial of risk-talking easier and leaves girls more vulnerable to peer pressure. With this constellation of effects, it is surprising that researchers are finding links between childhood sexual abuse and such behaviors as intravenous drug use, alcohol abuse, and precocious sexuality (Worth 1991).
Effect on children
Children who witness wife abuse are at risk of being assaulted themselves and of developing adjustment problems during childhood and adolescence. In a study of battered women presenting to the Institute of Legal Medicine in Bogota, Colombia, 74 percent of those who had children said that their children were present during the attack. In 49 percent of cases the children were also injured (Berenguer 1988). Of 80 women presenting to the Medico Forense of San Jose, Costa Rica, 40 percent said that their children were also beaten by their partner (Ugalde 1988). And in a representative survey of women in Texas, 33 percent of those abused during their lifetime had children who were abused by the same person (Teske and Parker 1983).
Perhaps even more significant than the physical injury that results from family violence is the effect it has on children's sense of security and their developing personalities. Two recent studies show that children who witness violence experience many of the same emotional and behavioral problems that abused children do, including depression, aggression, disobedience, nightmares, poor school performance, and somatic health complaints (Davis and Carlson 1987; Jaffe and others 1986). And evidence from Canada and the United States suggests that children who witness or experience violence are more likely to be abusive as adults (Stordeur and Stille 1989).
But violence may affect child survival in another, more subtle way. It is well established that female education is significantly and independently related to child survival (Blumberg 1989). What is not yet clear is how education affects child health. There is increasing evidence that schooling works not by imparting new knowledge or skills relating to health, but by eroding fatalism, improving women's self-confidence, and changing the balance of power in the family (Lindenbaum, Chakraborty, and Elias 1985; Levine and others 1987; Caldwell 1979). In the words of Peter Adamson (1988), "Education erodes resignation and substitutes for it a degree of confidence. an awareness of choice, a belief that decisions can be made, circumstances changed, life improved. "Using qualitative research techniques, Griffiths (1988) has identified some of the mechanisms through which maternal confidence and self-esteem affect child health. Her research in Cameroon, India, and Indonesia has demonstrated that mothers with higher self-esteem take a more assertive role in their child's feeding-they introduce weaning foods at the appropriate age, they take swifter action when a child is sick, and they persist in feeding even when a child refuses. Not surprisingly, more confident mothers have better nourished children.
If education is in fact a proxy for some intervening variable such as self-confidence or autonomy, anything that undermines confidence will affect child health. Acts of violence and society's tacit acceptance of them stand as constant reminders to women of their low worth. Where women's confidence and status are critical to achieving a development goal-such as improving child survival- violence, or the fear of it, will remain a powerful obstacle to progress.
New empirical data also link abuse of women by their husbands to the nutritional status of their children. In a census study of married women in three villages in rural Karnataka, India, qualitative and quantitative data indicated that inadequate payment of dowries and men's consumption of alcohol were the single greatest predictors of whether a wife would be beaten (Rao and Blach 1993). The children of women who were beaten were mot-e mal nourished and received less food than other equivalent children, a result the authors suggest may stem from the effects of wife beating on women's bargaining position m marriage.