In the 1970s, domestic (or family) violence, in particular spousal or intimate partner violence (IPV), was transformed from a private issue into a public concern that warrants intervention by the criminal justice system. Societal acceptance of family violence as a criminal matter has made demands on the police to change their traditional (or nonintervention) practices of handling victims and offenders of IPV, resulting in reforms that include proarrest policies. We provide a definition of domestic violence or IPV and describe its patterns, reviewing the history of domestic violence laws, traditional police practices in responding to it, and reforms in police response. Research on police responses to domestic violence, factors that influence victims’ calls to police, and the effectiveness of proarrest practices are presented, and remaining concerns in this area are discussed.
Definition and Extent of Domestic (Intimate Partner) Violence
Domestic violence is defined as ”the willful intimidation, assault, battery, sexual assault, or other abusive behavior perpetrated by one family member, household member, or intimate partner against another” (National Center for Victims of Crime 2005). It includes physical, sexual, and psychological abuse of a family member, as well as the destruction of property, which is why the term ”intimate partner abuse” is now more widely accepted than ”domestic violence.” Physical abuse commonly includes rape or sexual assaults on the victim (Johnson and Sigler 1997). In most cases of domestic violence, mental and physical abuse occur together. Verbal abuse includes threats regarding the woman’s welfare and the welfare of the children and degrading comments.
In most state laws criminalizing domestic violence, a charge of this offense requires that the parties are spouses, former spouses, cohabitating persons, or those who have resided together within the previous year, or persons who share a common child. A significant number of states also include dating relationships within the scope of domestic violence. Research confirms that a substantial amount of abuse occurs among persons who are not married to each other (for example, who are dating, cohabitating, ex-spouses, or ex-cohabitants [Erez 1986; Tjaden and Thoennes 2000]). Research has consistently found that the majority (more than 90%) of the victims are women. There is currently an academic debate as to the extent to which domestic violence is gendered, but the most careful research designs report it as being predominantly male offenders and female victims, and this is most pronounced when the measures include high levels of fear, injury, and abuse (Belknap and Melton 2005).
Domestic violence data are typically derived from calls to the police or national survey data. The severity of the injuries and the danger facing the victims of abuse are often high. Weapons are used in about a third of all domestic violence incidents. Almost a third of all women murdered in the United States are killed by their husbands, ex-husbands, or lovers, making IPV the largest cause of injury to women in the United States (Tjaden and Thoennes 2000)
Historical Background
Domestic violence appears to be a cultural universal; its historical roots are ancient and deep. In many societies the woman was essentially defined as her husband’s property; her sole purpose to satisfy his needs—bearing his children and tending to his household (Martin 1976). In medieval times, husbands had the power of life and death over their dependents and the right to unrestrained physical chastisement of the wife and other family members (Pleck 1987). Physical cruelty, including murder of the wife, was allowed as long as it was inflicted for disciplinary purposes (Davis 1971). Men killed their wives for reasons such as talking back, scolding and nagging, miscarrying, or sodomy (Martin 1976).
For ”family protection” purposes, the English common law provided husbands the right to chastise their wives ”moderately”; it excluded death (Blackstone 1987, 177). This law was brought to the American colonies, which allowed husbands to physically chastise their wives, as long as they did not use a stick larger than their thumb. Courts did not consider family violence incidents proper matters for court intervention, arguing that family arguments were best left inside the walls of the home. Exceptions to this rule were when the violence was excessive, was exercised merely to gratify ”bad passions,” or resulted in lasting injury. The economic and legal dependency of women on their husbands ”justified” the state’s nonintervention, which had an overriding interest in keeping the family intact. The sanctity of the family home and the maxim that ”a man’s home is his castle” were major reasons for a de facto decriminalization of wife abuse. They led to police treating spouse abuse differently than assaults between persons who were not intimate, resulting in policy of nonintervention in domestic violence cases.
The husband’s right to use necessary force to make the wife ”behave” and ”know her place” was not challenged in court until the end of the nineteenth century. At that time many of the legal restrictions on married women were lifted, and the right of the husband to chastise his wife was abolished. Yet, the belief that physical abuse in spousal relationships does not constitute a crime continued to guide the police in their responses. Until the 1970s, police officers treated domestic violence as a private matter, unsuitable for the intervention of the criminal justice system. At that time, the women’s movement began calling attention to the problem, and victim advocates initiated actions designed to bring about a change in police handling of domestic violence cases.
Police Attitudes and Changing Practices Concerning Domestic Violence
Until the 1980s, arrest in misdemeanor domestic violence cases was rarely made since police attached low priority to domestic violence. In the police culture, intervention in domestic situations was not viewed as ”real” police work; the police perceived intervention in it as unglamorous and unrewarding. Thus, police often ignored these calls, or if they did respond, treated it as a ”family dispute” where everyone needed to calm down.
The low level of enforcement of domestic violence incidents, however, was primarily attributed to legal requirements that the offense be committed in the presence of an officer, which technically barred officers who did not witness the abuse from making warrantless arrests. Other possible explanations include the erroneous police perception that domestic violence incidents are dangerous situations for the police, victim preferences against arrest, and possible police officers’ support or sympathy for the abusive male partner (Sherman 1992). This attitude was reinforced by the long-held belief that family conflict is a private matter and that taking action against abusers will hurt their families, particularly if they are economically dependent on the abuser. Female complainants were also viewed as uncooperative, making arrest (and prosecution) a waste of time in the eyes of the police (Parnas 1967).
This policy of avoiding making arrests received some professional attention in the 1960s, when social scientists and psychologists began to advocate mediation in ”family disturbance” incidents as the appropriate measures to handle IPV. Police officers received training in mediation, and many police departments established family crisis intervention units or police crisis teams that included social workers. This approach resulted in a further decrease in arrests in cities in which crisis intervention was practiced. Mediation resulted in keeping domestic violence out of the criminal justice system (Hirschel et al. 1992).
Mediation was soon rejected by both the police and women’s groups. For police officers, mediation seemed like social work rather than activities suitable for being considered police work. There was also no evidence to show that mediation was useful in the long run. Women’s groups objected to this approach because it ignored or underplayed the danger to women in abusive relationships. Mediation assumes equality of culpability between the parties to a dispute and fails to hold the offender accountable for his actions, inadvertently contributing to a dangerous escalation of the violence. In some jurisdictions (California, New York, and Connecticut) women’s groups filed suits against police departments on behalf of abused women whom the police failed to protect by arresting their abusers. The victims in these cases were awarded large sums of money for police negligence in protecting them. This litigation put pressure on the police to adopt a full enforcement policy in domestic violence. Arrest was now advocated as domestic violence began to be perceived as a criminal act, to be treated like assaults among nonfamily members. The pressure of litigation alone, however, could not resolve the problem of the ”in presence” requirement for warrantless misdemeanor arrests. Legislative action was necessary to provide officers with the power to arrest in misdemeanor domestic violence cases (Sherman 1992).
The preferred (or presumptive) arrest policy advocated slowly began to be transformed into laws in several states, and by 1984 the number of police departments that had adopted such policies had increased fourfold. Nonetheless, the actual implementation of this policy remained problematic, since there was no evidence to show that arrest in fact deters recidivism by abusive partners. And research still reports troubling responses by police to battered women (Erez and Belknap 1998).
Deterrent Effect of Arrest on Repeated Domestic Violence
Despite not being necessary to justify arrest in criminal offenses, the deterrent effect of arrest has become a central issue in attempts to convince policy makers to change the traditional police nonintervention practices in domestic violence cases. To provide evidence that arrest deters bat-terers, a scientific study of the effect of arrest on domestic violence recidivism rates was needed. In the early 1980s in Minneapolis, Sherman and Berk (1984) conducted an experimental design study comparing the effects of the traditional responses of the police (separation of the parties and mediation or advising) and arrest on subsequent battering by intimates (”the Minneapolis experiment”). Analysis of the data showed that arrest was more effective than the other two responses in deterring repeat abuse.
The Minneapolis experiment received national attention and was instrumental in promoting arrest as the preferred response to domestic violence cases across the country. By 1989, about three-quarters of the large city police departments had adopted a mandatory arrest policy and more than 80% a preferred arrest policy (Hirschel et al. 1992). Predictably, the national arrest rate for misdemeanor domestic violence rose by 70% from 1985 to 1989 (Sherman 1992). However, this did not amount to a full compliance with proarrest policies. Research shows that despite official proarrest policies by many police departments, the arrest rate continues to be low (for example, Buzawa and Buzawa 1996; Ferraro 1989).
Several factors may account for this low compliance: persistent negative attitudes of the police concerning the seriousness of the abuse (Belknap 1992) or the appropriateness of intervention in family matters, identification with the abuser, fear of being sued in civil court for wrongful arrest (Hirschel et al. 1992), and the erroneous perception that intervention is highly dangerous to the police. The perception about higher probability of injury or death to the officers who intervene in domestic violence has been a part of police folklore for a long time and was often promoted in crisis intervention training to convince recruits who were reluctant to adopt the mediation approach. This perception was also supported by misinterpretation of ”disturbance calls” data, which grouped family violence with all other types of disturbances, such as fights in bars (Garner and Clemmer 1986). While an in-depth analysis of police injury and death rates has raised serious doubts about such danger to officers, some studies continue to suggest that in some locations domestic calls may still constitute the most dangerous category for police, both in assault and injury (Hirschel et al. 1992).
To overcome remaining resistance to the preferred arrest policy by some police officers and departments and to provide further evidence of the superiority of the arrest response, the National Institute of Justice (NIJ) funded replications of the Minneapolis experiment. These six studies, known collectively as the NIJ’s Spouse Assault Replication Program (SARP) field experiments, were carried out between 1981 and 1991 by police departments and research teams in Omaha, Nebraska,
Charlotte, North Carolina, Milwaukee, Wisconsin, Colorado Springs, Colorado, Dade County, Florida, and Atlanta, Georgia. The combined results of five studies (all except Atlanta; see Maxwell, Garner, and Fagan 2001) show that arresting batterers was related to reduced subsequent aggression against female intimate partners, although not all comparisons met the standard level of statistical significance. The impact of arrest on repeat offending (that is, the deterrent effect of arrest) was also found to be modest when compared to such factors as the bat-terer’s prior criminal record or age. It showed that regardless of whether the bat-terer was arrested, more than half of the suspects committed no subsequent criminal offense against the original victim during the follow-up period, and a minority of suspects continued to commit intimate partner violence regardless of whether they were arrested, counseled, or temporarily separated from their partner (Maxwell, Garner, and Fagan 2001).
Legal experts, feminist scholars, social activists, and researchers have criticized the attention to the deterrent effect of arrest on batterers and the interpretation of the Minneapolis experiment and its replications. Practitioners and women’s advocates argue that arrest alone is ineffective in halting the long-term expected progression of violence by socially marginal offenders. Research has documented that domestic violence escalates in spousal relationships whether or not the abuser is arrested (Frisch 1992). Social activists argue that the heavy emphasis on arrest as a panacea for domestic violence detracts from the role of community attitudes and practices in determining the scope and nature of the problem. The preoccupation with pro-arrest policies results in focusing on the individual rather than acknowledging societal factors that perpetuate the dependency of women on batterers (Ferraro 1989). They also point out the need for coordination among the police, the judiciary, and social services in responding to domestic violence, for additional resources to pursue IPV cases through the legal system, and for higher probability of abusers’ prosecution (Hirschel et al. 1992).
Feminists have argued that the experiments’ effects were interpreted from the viewpoint of the abuser and ignored the perspective of the battered woman. For instance, the importance of the employment status of victims has not been taken into account; employed women are more able to successfully leave battering relationships than their unemployed counterparts. Generally, the Minneapolis experiment and its replications have centered on police response almost exclusively in terms of the arrest versus mediation decision. Other police actions, such as referrals made by the police for battered victims and their offenders for programs, shelters, and rights are critical and need to be addressed. The research on the deterrent effect of proarrest policies has also limited the definition of ”success” to whether arresting batterers reduces their recidivism rates. This view ignores other influences that arrest might have, such as providing victims and their children an opportunity to escape, as well as communicating to offenders and victims and their children that the batterers’ behavior is reprehensible.
Mandatory or presumed arrest policies and practices have been criticized. Research shows that many victims would prefer to resolve the incident without an arrest (Kingsnorth and Macintosh 2004). When victims do call the police, many merely want the violence to end, not to have their abuser arrested. In some cases, police are called to the scene by a third party rather than the victim. The majority of IPV victims do not even call the police (Buzawa and Buzawa 1996; Hoyle and Sanders 2000). Fear of reprisal, perceived social stigma, and a belief that nothing can be accomplished by reporting all reduce victim willingness to call the police.
Some researchers have attributed the drop in calls to the police for assistance in domestic violence cases following the passage of mandatory arrest laws to women’s fears that their abusers will be arrested (Felders 2001). From the victim’s perspective, the costs of involving the police (for example, losing the abuser’s financial support or fear of loneliness or of future problems such as retaliatory violence) outweigh the benefits. In many cases, however, arrest accomplishes what the victim wanted in terms of changes in the offender’s behavior, and prosecution is not necessary. A major criticism of mandatory arrest policies is that they deprive women of choice. It is argued that despite victims’ emotional involvement and trauma, they are usually in a better position than patrol officers to determine the likely impact of an offender’s arrest on their safety.
Finally, the implementation of pro-arrest policies led to the unprecedented large numbers of women arrested for domestic violence, even though they were not the primary aggressor. Most of these women appeared to be the victims, but police arrested them based on the male offender claiming they were abusive or the police simply arresting both the man and woman (dual arrests) when a woman resisted the abuse in any way (see Belknap and Melton 2005).
Conclusion
The gradual changes in societal attitudes toward spousal abuse and recognition of its criminal nature have led to a reevaluation of state law concerning domestic violence and reforms in police response to it. Currently, most police departments throughout the country adopt either preferred or mandatory arrest policies, although this reform did not result in a substantial increase in arrest rates. Police response to domestic violence and the amount of discretion they may or should use in making arrests in misdemeanor incidents remains an open question. The magnitude of the IPV problem justifies attention to the deterrent capability of arrest from a victim safety perspective. Arrest, which is mainly a response to criminal conduct, also serves as the gateway to the justice system. On the other hand, preoccupation with its deterrence capability ignores the fact that arrest alone may not be a panacea and that policy makers need to explore other avenues to handle this persistent social problem.
