The Better Angels of Our Nature: Why Violence Has Declined
The other major category of violence against women has been called wife-beating, battering, spousal abuse, intimate partner violence, and domestic violence. The man uses physical force to intimidate, assault, and in extreme cases kill a current or estranged wife or girlfriend. Usually the violence is motivated by sexual jealousy or a fear that the woman will leave him, though he may also use it to establish dominance in the relationship by punishing her for acts of insubordination, such as challenging his authority or failing to perform a domestic duty.71
Domestic violence is the backstop in a set of tactics by which men control the freedom, especially the sexual freedom, of their partners. It may be related to the biological phenomenon of mate-guarding.72 In many organisms in which males invest in their offspring and females have opportunities to mate with other males, the male will follow the female around, try to keep her away from rival males, and, upon seeing signs that he may have failed, attempt to copulate with her on the spot. Human practices such as veiling, chaperoning, chastity belts, claustration, segregation by sex, and female genital cutting appear to be culturally sanctioned mate-guarding tactics. As an extra layer of protection, men often contract with other men (and sometimes older female kin) to recognize their monopoly over their partners as a legal right. Legal codes in the civilizations of the Fertile Crescent, the Far East, the Americas, Africa, and northern Europe spelled out almost identical corollaries of the equation of women with property.73 Adultery was a tort against the husband by his romantic rival, entitling him to damages, divorce (with refund of the bride-price), or violent revenge. Adultery was always defined by the marital status of the woman; the man’s marital status, and the woman’s own preferences in the matter, were immaterial. Well into the first decades of the 20th century, the man of the house was entitled by law to “chastise” his wife.74
In Western countries the 1970s saw the repeal of many laws that had treated women as possessions of their husbands. Divorce laws became more symmetrical. A man could no longer claim justifiable provocation when he killed his adulterous wife or her lover. A husband could no longer forcibly confine his wife or prevent her from leaving the house. And a woman’s family and friends were no longer guilty of the crime of “harboring” if they gave sanctuary to a fleeing wife.75 Most parts of the United States now have shelters in which women can escape from an abusive partner, and the legal system has recognized their right to safety by criminalizing domestic violence. Police who used to stay out of “marital spats” are now required by the laws of a majority of states to arrest a spouse if there is probable cause of abuse. In many jurisdictions prosecutors are obligated to seek protective orders that keep a potentially abusive spouse away from his home and partner, and then to prosecute him without the option of dropping the case, whether the victim wants it to proceed or not.76 Originally intended to rescue women who were trapped in a cycle of abuse, apology, forgiveness, and reoffending, the policies have become so intrusive that some legal scholars, such as Jeannie Suk, have argued that they now work against the interests of women by denying them their autonomy.
Attitudes have changed as well. For centuries wife-beating was considered a normal part of marriage, from the quip of the 17th-century playwrights Beaumont and Fletcher that “charity and beating begins at home,” to the threat of the 20th-century bus driver Ralph Kramden, “One of these days, Alice . . . POW, right in the kisser.” As recently as 1972, the respondents to a survey of the seriousness of various crimes ranked violence toward a spouse 91st in a list of 140. (Respondents in that survey also considered the selling of LSD to be a worse crime than the “forcible rape of a stranger in the park.”)77 Readers who distrust survey data may be interested in an experiment conducted in 1974 by the social psychologists Lance Shotland and Margaret Straw. Students filling out a questionnaire overheard an argument break out between a man and a woman (in reality, actors hired by the experimenters). I will let the authors describe the experimental method:After approximately 15 sec of heated discussion, the man physically attacked the woman, violently shaking her while she struggled, resisting and screaming. The screams were loud piercing shrieks, interspersed with pleas to “get away from me.” Along with the shrieks, one of two conditions was introduced and then repeated several times. In the Stranger Condition the woman screamed, “I don’t know you,” and in the Married Condition, “I don’t know why I ever married you.”78
Most of the students ran out of the testing room to see what the commotion was about. In the condition in which the actors played strangers, almost two-thirds of the students intervened, usually by approaching the couple slowly and hoping they would stop. But in the condition in which the actors played husband and wife, fewer than a fifth of the students intervened. Most of them did not even pick up a phone in front of them with a sticker that listed an emergency number for the campus police. When interviewed afterward, they said it was “none of their business.” In 1974 violence that was considered unacceptable between strangers was considered acceptable within a marriage.
The experiment almost certainly could not be conducted today because of federal regulations on research with human subjects, another sign of our violence-averse times. But other studies suggest that people today are less likely to think that a violent attack by a man on his wife is none of their business. In a 1995 survey more than 80 percent of the respondents deemed domestic violence a “very important social and legal issue” (more important than children in poverty and the state of the environment), 87 percent believed that intervention is necessary when a man hits his wife even if she is not injured, and 99 percent believe that legal intervention is necessary if a man injures the wife. 79 Surveys that ask the same questions in different decades show striking changes. In 1987 only half of Americans thought it was always wrong for a man to strike his wife with a belt or stick; a decade later 86 percent thought it was always wrong.80 Figure 7–12 shows the statistically adjusted results of four surveys that asked people whether they approved of a husband slapping a wife. Between 1968 and 1994 the level of approval fell by half, from 20 to 10 percent. Though men are more likely to condone domestic violence than women, the feminist tide carried them along as well, and the men of 1994 were less approving than the women of 1968. The decline was seen in all regions of the country and in samples of whites, blacks, and Hispanics alike.
What about domestic violence itself? Before we look at the trends, we have to consider a surprising claim: that men commit no more domestic violence than women. The sociologist Murray Straus has conducted many confidential and anonymous surveys that asked people in relationships whether they had ever used violence against their partners, and he found no difference between the sexes.81 In 1978 he wrote, “The old cartoons of the wife chasing her husband with a rolling pin or throwing pots and pans are closer to reality than most (and especially those with feminist sympathies) realize.”82 Some activists have called for greater recognition of the problem of battered men, and for a network of shelters in which men can escape from their violent wives and girlfriends. This would be quite a twist. If women were never the victims of a gendered category of violence called “wife-beating,” but rather both sexes have always been equally victimized by “spouse-beating,” it would be misleading to ask whether wife-beating has declined over time as a part of the campaign to end violence against women.
To make sense of the survey findings, one has to be careful about what one means by domestic violence. It turns out that there really is a distinction between common marital spats that escalate into violence (“the conversations with the flying plates,” as Rodgers and Hart put it) and the systematic intimidation and coercion of one partner by another.83 The sociologist Michael Johnson analyzed data on the interactions between partners in violent relationships and discovered a cluster of controlling tactics that tended to go together. In some couples, one partner threatens the other with force, controls the family finances, restricts the other’s movements, redirects anger and violence against the childre
n or pets, and strategically withholds praise and affection. Among couples with a controller, the controllers who used violence were almost exclusively men; the spouses who used violence were almost entirely women, presumably defending themselves or their children. When neither partner was a controller, violence erupted only when an argument got out of hand, and in those couples the men were just a shade more prone to using force than the women. The distinction between controllers and squabblers, then, resolves the mystery of the genderneutral violence statistics. The numbers in violence surveys are dominated by spats between noncontrolling partners, in which the women give as good as they get. But the numbers from shelter admissions, court records, emergency rooms, and police statistics are dominated by couples with a controller, usually the man intimidating the woman, and occasionally a woman defending herself. The asymmetry is even greater with estranged partners, in which it is the men who do most of the stalking, threatening, and harming. Other studies have confirmed that chronic intimidation, serious violence, and maleness tend to go together.84
FIGURE 7–12. Approval of husband slapping wife in the United States, 1968–1994
Source: Graph from Straus et al., 1997.
So has anything changed over time? With the small stuff—the mutual slapping and shoving—perhaps not.85 But with violence that is severe enough to count as an assault, so that it turns up in the National Crime Victimization Survey, the rates have plunged. As with estimates of rape, the numbers from a victimization study can’t be treated as exact measures of rates of domestic violence, but they are useful as a measure of trends over time, especially since the new concern over domestic violence should make the recent respondents more willing to report abuse. The Bureau of Justice provides data from 1993 to 2005, which are plotted in figure 7–13. The rate of reported violence against women by their intimate partners has fallen by almost two-thirds, and the rate against men has fallen by almost half.
The decline almost certainly began earlier. In Straus’s surveys, women in 1985 reported twice the number of severe assaults by their husbands as did the women in 1992, the year the federal victimization data begin.86
What about the most extreme form of domestic violence, uxoricide and mariticide? To a social scientist, the killing of one intimate partner by another has the great advantage that one needn’t haggle over definitions or worry about reporting biases: dead is dead. Figure 7–14 shows the rates of killing of intimate partners from 1976 to 2005, expressed as a ratio per 100,000 people of the same sex.
FIGURE 7–13. Assaults by intimate partners in the United States, 1993–2005
Source: Data from U.S. Bureau of Justice Statistics, 2010.
FIGURE 7–14. Homicides of intimate partners in the United States, 1976–2005
Sources: Data from U.S. Bureau of Justice Statistics, 2011, with adjustments by the Sourcebook of Criminal Justice Statistics Online (http://www.albany.edu/sourcebook/csv/t31312005.csv). Population figures from the U.S. Census.
Once again, we see a substantial decline, though with an interesting twist: feminism has been very good for men. In the years since the ascendancy of the women’s movement, the chance that a man would be killed by his wife, ex-wife, or girlfriend has fallen sixfold. Since there was no campaign to end violence against men during this period, and since women in general are the less homicidal sex, the likeliest explanation is that a woman was apt to kill an abusive husband or boyfriend when he threatened to harm her if she left him. The advent of women’s shelters and restraining orders gave women an escape plan that was a bit less extreme.87
What about the rest of the world? Unfortunately it is not easy to tell. Unlike homicide, definitions of rape and spousal abuse are all over the map, and police records are misleading because any change in the rate of violence against women can be swamped by changes in the willingness of women to report the abuse to the police. Adding to the confusion, advocacy groups tend to inflate statistics on rates of violence against women and to hide statistics on trends over time. The U.K. Home Office administers a crime victimization survey in England and Wales, but it does not present data on trends in rape or domestic violence.88 But when the data from separate annual reports are aggregated, as in figure 7–15, they show a dramatic decline in domestic violence, similar to the one seen in the United States. Because of differences in how domestic violence is defined and how population bases are tabulated, the numbers in this graph are not commensurable with those of figure 7–13, but the trends over time are almost the same. It is safe to guess that similar declines have taken place in other Western democracies, because domestic violence has been a concern in all of them.
FIGURE 7–15. Domestic violence in England and Wales, 1995–2008
Sources: Data from British Crime Survey, U.K., Home Office, 2010. Data aggregated across years by Dewar Research, 2009. Population estimates from the U.K. Office for National Statistics, 2009.
Though the United States and other Western nations are often accused of being misogynistic patriarchies, the rest of the world is immensely worse. As I mentioned, surveys of domestic violence in the United States that are broad enough to include minor acts of shoving and slapping show no difference between men and women; the same is true of Canada, Finland, Germany, the U.K., Ireland, Israel, and Poland. But that gender neutrality is a departure from the rest of the world. The psychologist John Archer looked at sex ratios in surveys of domestic violence in sixteen countries and found that in the non-Western ones—India, Jordan, Japan, Korea, Nigeria, and Papua New Guinea—the men do more of the hitting.89
The World Health Organization recently published a hodgepodge of rates of serious domestic violence from forty-eight countries.90 Worldwide, it has been estimated that between a fifth and a half of all women have been victims of domestic violence, and they are far worse off in countries outside Western Europe and the Anglosphere.91 In the United States, Canada, and Australia, fewer than 3 percent of women report that their partners assaulted them in the previous year, but the reports from other countries are an order of magnitude higher: 27 percent in a Nicaraguan sample, 38 percent in a Korean sample, and 52 percent in a Palestinian sample. Attitudes toward marital violence also show striking differences. About 1 percent of New Zealanders and 4 percent of Singaporeans say that a husband has the right to beat a wife who talks back or disobeys him. But the figures are 78 percent for rural Egyptians, up to 50 percent for Indians in Uttar Pradesh, and 57 percent for Palestinians.
Laws on violence against women also show a lag from the legal reforms of W estern democracies.92 Eighty-four percent of the countries of Western Europe have outlawed or are planning to outlaw domestic violence, and 72 percent have done so for marital rape. Here are the respective figures for other parts of the world: Eastern Europe, 57 and 39 percent; Asia and the Pacific, 51 and 19 percent; Latin America, 94 and 18 percent; sub-Saharan Africa, 35 and 12.5 percent; Arab states, 25 and 0 percent. On top of these injustices, sub-Saharan Africa and South and Southwest Asia are host to systematic atrocities against women that are rare or unheard of in the 21st-century West, including infanticide, genital mutilation, trafficking in child prostitutes and sex slaves, honor killings, attacks on disobedient or under-dowried wives with acid and burning kerosene, and mass rapes during wars, riots, and genocides.93
Is the difference in violence against women between the West and the rest just one of the many wholesome factors that are bundled together in the Matthew Effect—democracy, prosperity, economic freedom, education, technology, decent government? Not entirely. Korea and Japan are affluent democracies but have more domestic violence against women, and several Latin American countries that are far less developed appear to have more equal sex ratios and lower absolute rates. This leaves some statistical wiggle room to look for the differences across societies that make women safer, holding affluence constant. Archer found that countries in which women are better represented in government and the professions, and in which they earn a larger proportion of earned income, are le
ss likely to have women at the receiving end of spousal abuse. Also, cultures that are classified as more individualistic, where people feel they are individuals with the right to pursue their own goals, have relatively less domestic violence against women than the cultures classified as collectivist, where people feel they are part of a community whose interests take precedence over their own.94 These correlations don’t prove causation, but they are consistent with the suggestion that the decline of violence against women in the West has been pushed along by a humanist mindset that elevates the rights of individual people over the traditions of the community, and that increasingly embraces the vantage point of women.
Though elsewhere I have been chary about making predictions, I think it’s extremely likely that in the coming decades violence against women will decrease throughout the world. The pressure will come both from the top down and from the bottom up. At the top, a consensus has formed within the international community that violence against women is the most pressing human rights problem remaining in the world.95 There have been symbolic measures such as the International Day for the Elimination of Violence Against Women (November 25) and numerous proclamations from bully pulpits such as the United Nations and its member governments. Though the measures are toothless, the history of denunciations of slavery, whaling, piracy, privateering, chemical weaponry, apartheid, and atmospheric nuclear testing shows that international shaming campaigns can make a difference over the long run.96 As the head of the UN Development Fund for Women has noted, “There are now more national plans, policies, and laws in place than ever before, and momentum is also growing in the intergovernmental arena.”97