Fostered, if not mandated by the Violence Against Women Act (VAWA) of 1994, Alaska, Arizona, Colorado, Connecticut, District of Columbia, Iowa, Kansas, Louisiana, Maine, Mississippi, Nevada, New Jersey, New York, Ohio, Oregon, Rhode Island, South Carolina, South Dakota, Utah, Virginia, Washington, and Wisconsin all passed laws making arrest mandatory in domestic violence cases. In addition Arkansas, California, Florida, Massachusetts, Michigan, Montana, North Dakota, and Tennessee passed "preferred arrest" laws that amounted to the same thing.
In the year 2000 I began looking at what impact Colorado's mandatory arrest law had in Colorado Springs, one of seven cities in the United States where the mandatory arrest policy had been tested prior to becoming law. Initially I plotted police data for the ten-year period from 1990 (before the law was passed) to 1999 (five years after mandatory arrest became law). That plot was later expanded to the fifteen-year period from 1990 to 2004 and is shown below.
There is a spike in 911 calls for domestic disturbances in 1994. But by 1995 the word about the consequences of calling 911 in a "domestic" had spread like wildfire and 911 domestic disturbance calls dropped dramatically, and kept dropping as shown by the red line in the above plot. Clearly, the draconian police response mandated by the 1994 DV laws primarily acts to deter citizens from calling 911. Subsequent research by others found the same result in at least three other cities.
My initial query was to see how much arrests for simple assault (roughly 70-75% of DV arrests are for simple assault) increased after 1994. From the plot (gold line) it is apparent these arrests roughly doubled in 1995 but then began to drop and by 2004 were no more common than a linear increase due to population would predict. Thus, the major impact of the DV laws is clearly to deter citizens from calling police in a domestic disturbance.
© 2007 The New York Times
Reproduced under the Fair Use exception of 17 USC § 107 for noncommercial, nonprofit, and educational use.
August 7, 2007 Two decades ago, in an effort to curb domestic violence, states began passing "mandatory arrest" laws. Police officers responding to a call for help would no longer need to determine whether one person was truly violent or out of control; every time someone reported abuse, the police would simply be required to make an arrest.
It seemed like a good tactic at least to people who work with victims of domestic violence. (Police officers tended to be less enthusiastic, because they prefer to make arrests at their own discretion.) Arrests would immediately stop the violence and might discourage abusers from further acts of abuse.
But 20 years later, it seems the mandatory arrest laws are having an unintended, deadly side effect. The number of murders committed by intimate partners is now significantly higher in states with mandatory arrest laws than it is in other states.
Support for the laws began in 1984, after a federal district court in Connecticut ruled that the police had inadequately protected a woman whose husband had brutally assaulted her. State lawmakers decided they needed more control over how local police departments enforced restraining orders against abusers and intervened in incidents of violence. One way to get that control was to dictate how the police should respond in each case.
A small but influential study of police responses to domestic violence calls, conducted by criminologists in Minnesota in the early 1980s, found that arrests were the most effective strategy for reducing future violence. Now, 22 states and the District of Columbia have laws that mandate or at least strongly recommend that everyone accused of domestic abuse be arrested.
What the laws did not take into account was that eventually the victims of violence would come to realize that if they called the police, their abuser would certainly be arrested. And over the years, it turns out, that realization seems to have led victims to contact the police less.
I recently conducted my own study of mandatory arrest laws by comparing the rates of murders by intimate partners before and after the laws went into effect. Intimate partner homicides have generally decreased in the past 20 years, perhaps because greater awareness of the problem of domestic violence has led to the creation of more resources for victims. But in states with mandatory arrest laws, the homicides are about 50 percent higher today than they are in states without the laws.
The mandatory arrest laws were intended to impose a cost on abusers. But because of psychological, emotional and financial ties that often keep victims loyal to their abusers, the cost of arrest is easily transferred from abusers to victims. Victims want protection, but they do not always want to see their partners put behind bars.
In some cases, victims may favor an arrest, but fear that their abusers will be quickly released. And many victims may avoid calling the police for fear that they, too, will be arrested for physically defending themselves. The possibility of such "dual arrests" is most worrisome for victims who have children at home.
The situation is different in incidents in which abuse is suffered by people who are not intimate partners children, for example. The certainty of arrest does nothing to deter the reporting of such cases, usually by teachers, doctors or other third parties. In fact, my research shows that in states with mandatory arrest laws there are fewer murders of non-intimate-partner family members than there are in states without the laws.
Despite two decades of increased public awareness, domestic violence remains a serious problem. Arresting abusers is often desirable, as are efforts to educate the police about domestic violence and effective intervention and to provide treatment and support for victims. But it makes no sense to keep following a strategy that discourages victims from reporting abuse.
Radha Iyengar is a fellow in health policy research at Harvard.