Abstract
Persons under certain domestic violence restraining orders in California are required to surrender any firearms in their possession within 24 hours of service. The California Department of Justice funded a pilot program in which Sheriff’s Offices in two counties developed a system for better enforcing the firearm surrender requirement. As part of a larger process evaluation, 17 restraining order recipients were interviewed about their experiences with and feelings about the removal of firearms from their abusers. Most women surveyed wanted firearms removed and felt safer as a result of their removal. Implications of the findings are discussed.
Introduction
The Federal Bureau of Investigation (FBI) reports that 1,360 individuals were murdered by a current or former intimate partner in 2009 in the United States, though there were likely to be hundreds more due to incomplete reporting to the Supplemental Homicide Reports program and unsolved murders (FBI, 2010). Most intimate partner homicides (IPH) are preceded by a long history of non-fatal violence, and most occur when a male abuser shoots a female victim (Campbell et al., 2003; Fox & Zawitz, 2006). A woman is at highest risk for IPH immediately after she attempts to leave a partner who has been physically abusive (Campbell et al., 2003; Wilson & Daly, 1993). An abuser’s access to a firearm is associated with a fivefold increased risk of the female partner being murdered by her abuser (Campbell et al., 2003).
During the past three decades, rates of IPH have declined steadily in the U.S. (Browne & Williams, 1993; Fox & Zawitz, 2006; Jennings & Piquero, 2008; Wells & DeLeon-Granados, 2004). There has been limited research on whether specific policies and programs can be credited for this downward trend, and findings have been equivocal. Research by Dugan, Rosenfeld, and Nagin (2003) suggests that interventions that enable victims to separate from their abusive partners tend to reduce IPH, but certain interventions have the potential to provoke retaliatory violence by threatening an abuser’s sense of power and control. For example, policies allowing warrantless arrests for violations of domestic violence restraining orders were found to be associated with reductions in IPH, but aggressive prosecution of domestic violence cases was associated with higher levels of IPH (Dugan et al., 2003).
A domestic violence restraining order (DVRO) is intended to prevent further abuse by separating a victim from her abuser. The potential protective effects of DVROs may be enhanced by laws that prohibit DVRO defendants from purchasing and possessing firearms. 1 Although research suggests that enforcement of these laws is lacking (Frattaroli & Teret, 2006; Moracco, Clark, Espersen, & Bowling, 2006; Seave, 2006; Webster et al., 2010), firearm prohibitions for DVRO defendants have been associated with reductions in IPH (Bridges, Tatum, & Kunselman, 2008; Vigdor & Mercy, 2003, 2006; Zeoli & Webster, 2010).
Firearm removal from batterers is widely supported by domestic violence experts, advocates, law enforcement, and the general public (Gwinn, 2006; Moracco et al., 2006; Seave, 2006; Sorenson, 2006a, 2006b). Mandating the removal of guns when serving a DVRO was among the recommendations arising from a 2007 meeting of the International Association of Chiefs of Police on gun violence. In addition, according to a community-based survey of 522 adults in California, more than three quarters of California adults support the removal of guns from batterers (Sorenson, 2006b).
Despite this wide support, it is important to consider the possibility that firearm removal from a batterer in a situation in which two major risk factors for IPH are present—a victim separating from an abuser and an abuser’s access to a firearm—could potentially increase risk to the victim, law enforcement, and even to the abuser. The potential for retaliatory violence in response to firearm removal may be particularly great given the central role of firearms in the lives of some gun owners. It is believed that efforts to routinely remove firearms from abusers proscribed from possessing firearms due to a DVRO or criminal convictions for domestic violence are very rare. Furthermore, to our knowledge, no one has surveyed individuals most directly affected by firearm removal—the victims.
Victims’ experiences with and perceptions about DVROs in general are varied. Some studies document a high level of satisfaction with restraining orders among recipients (Harrell & Smith, 1996; Horton, Simonidis, & Simonidis, 1987; Kaci, 1994; Keilitz, Hannaford, & Efkeman, 1997), and obtaining a restraining order appears to impart a sense of empowerment to many victims (Cattaneo & Goodman, 2010; Chaudari & Daly, 1991; Fischer & Rose, 1995; Ptacek, 1999). Victims have also reported feeling safer after obtaining a restraining order (Harrell, Smith, & Newmark, 1993; Keilitz et al., 1997; Logan & Walker, 2009; Ptacek, 1999). Nonetheless, fear of retaliation by the abuser has been documented among abuse victims seeking restraining orders (Fischer & Rose, 1995; Ptacek, 1999), as well as among victims who utilize other criminal justice resources (Erez & Belknap, 1998; Felson, Messner, Hoskin, & Deanne, 2002; Hare, 2010). Moreover, at least one study found that domestic violence victims who obtained a restraining order felt less safe as a result (Erez & Belknap, 1998).
In California, the site of the current study, all persons who are under a DVRO must surrender their firearms to a local law enforcement agency or sell them to a licensed dealer within 24 hours of the restraining order being served. Firearm relinquishment in these cases must be immediate at the request of a law enforcement officer. As in other states, the extent to which firearm relinquishment actually occurs in California is questionable at best. A 2005 report to the California Attorney General found that none of the law enforcement agencies in the 10 counties studied had a policy to proactively enforce firearm prohibitions on civil DVROs (Attorney General’s Task Force on Local Criminal Justice Responses to Domestic Violence, 2005; Seave, 2006).
In response to this evidence, the California Department of Justice funded a pilot program in which the Sheriff’s Offices in two counties developed a system for better enforcing the firearm surrender requirement. This is one of the first attempts to proactively implement and enforce firearm prohibitions among restrained persons (Frattaroli, Claire, Vittes, Webster, & Wintemute, 2011). Thus, information about the impact of the program from the victims’ point of view is particularly important.
We report findings from pilot survey interviews with restraining order recipients about their experiences with firearm violence and the removal of firearms from their abusers. We were particularly interested in whether firearm removal from restrained persons provoked retaliatory violence and whether it increased or decreased victims’ feelings of safety.
Methods
Detectives assigned to the pilot program within the Sheriff’s Offices in Butte and San Mateo Counties received a copy of each new DVRO shortly after it was issued and checked each respondent against the state’s Automated Firearm System (a database of all handgun and assault weapon transfers taking place in California). Sheriff’s deputies also gained knowledge about firearm possession by restrained people from the declarations written by victims as part of the restraining order applications and through personal conversations with victims. After or at the time the restraining order was served, local police or a sheriff’s deputy informed the respondent that he must surrender all of his firearms as required by state law. Additional information on the pilot program can be found in this report (see Wintemute, Frattaroli, Vittes, Claire, & Webster, 2011).
Procedure
To examine the knowledge and perspectives of the victims involved in the program, we sought interviews with women who had obtained a temporary restraining order in the two counties. Unfortunately, we were unable to recruit women from Butte County, as the Sheriff’s Office reported that they did not have usable contact information for petitioners involved in the intervention.
A woman was eligible to participate in the survey if she was at least 18 years old, spoke English or Spanish, and received a civil temporary restraining order against a male current or former intimate partner in San Mateo County between October 2007 and June 2010. We only interviewed a woman if sheriff’s deputies identified firearms in their review of the restraining order or if law enforcement officers confiscated firearms from her abuser (i.e., the restrained person). We chose to limit the sample to victims of male-on-female domestic violence because about 80% of the protected persons on DVROs in California are women (Sorenson & Shen, 2005), men are substantially more likely than women to own and use firearms (Cook & Ludwig, 1997), and women are more likely than men to be killed by an intimate partner (Fox & Zawitz, 2006). We also limited our sample to restraining order recipients who were at least 18 years old given that almost all (99.3%) restraining orders in California are sought by adults, and research suggests that restraining orders among people below 18 years of age differ qualitatively from those among adults (Sorenson & Shen, 2005).
The research team identified 89 women from San Mateo County who met the stated eligibility criteria for participation in the survey. A sheriff’s deputy made the initial contact with the eligible participants to provide them with a brief overview of the study and to ask whether they were willing to be contacted by one of our interviewers. A female interviewer contacted the women who agreed to participate and who provided a telephone number. The interviewer obtained full verbal consent from each participant before proceeding with the interview. Study participants were given a US$25 gift card to Target or Amazon.com for participating in the 25-minute interview.
Interviews were conducted by telephone between October 19, 2009 and June 15, 2010 by two female interviewers experienced in interviewing victims of intimate partner violence. An average of 11 months elapsed between the issuance of the index temporary restraining order (i.e., the restraining order that prompted firearm removal) and the interview. Although interviews were offered in English and Spanish, all participants chose to be interviewed in English. None of the women terminated the interview prematurely.
Measures
The survey instrument included questions about demographic and relationship characteristics of both the protected (i.e., victim) and restrained persons (e.g., race, level of education, and marital and cohabitation status before and after obtaining the restraining order); the abuse that transpired in the 6 months before and 6 months after getting the restraining order; and about the restrained person’s guns (e.g., number, type, how the victim knew about them); the extent to which the judge and law enforcement officers inquired about the presence of guns; and whether the guns were relinquished. The interviewers also asked questions to assess the victims’ level of comfort with firearm removal by law enforcement, including whether they wanted to have the guns removed, how firearm removal (or lack thereof) affected their feelings of safety, and how the restrained person responded or might have responded to having to surrender his firearms. Finally, victims were asked whether they obtained a final restraining order after the temporary order expired, whether the restrained person requested that his guns be returned to him, and whether they were returned.
In addition to the survey data, we consulted data from logs kept by the San Mateo County Sheriff’s Office to compare the survey participants’ responses with law enforcement data. These logs included information about the number of firearms identified by law enforcement and surrendered by the restrained person and whether the survey participants obtained a final restraining order after the expiration of the temporary order.
Subject Recruitment and Participation
As described previously, recruitment occurred in two stages. Therefore, we calculated first stage, second stage, and overall noncontact and refusal rates. Of the 89 women identified by the research team as eligible to participate, a sheriff’s deputy made contact with 48 women. Six of the 48 women who were contacted refused to have their names and telephone numbers passed on to researchers. This resulted in a first-stage noncontact rate of 46.1% (41 out of 89) and a first stage refusal rate of 12.5% (6 out of 48). Three of the 42 women who agreed to have their names and telephone numbers given to researchers did not meet all of the eligibility criteria. Thus, the interviewers were provided with the names and telephone numbers of the remaining 39 women. The interviewers reached 21 of these 39 women for a second-stage noncontact rate of 46.2% (18 out of 39). Three of the 21 women contacted by the interviewers refused to participate resulting in a second-stage refusal rate of 14.3%. Combining the noncontact and refusal rates for the first and second stages resulted in an overall noncontact rate of 66.3% (59 out of 89) and an overall refusal rate of 18.8% (9 out of 48). Thus, 17 interviews were included in the analyses.
Data Analysis
The purpose of the pilot study was to provide descriptive information about women’s experiences with and perceptions about firearm removal by law enforcement. Thus, we provide descriptive data on the 17 participants below.
Findings
Demographic and Relationship Characteristics
Sociodemographic and relationship characteristics of the 17 protected and restrained persons are shown in Tables 1 and 2. For both protected and restrained persons, nearly half were White, most had at least a high school education, and most were employed. Most protected and restrained persons who had been in an intimate relationship or who were cohabiting before obtaining the temporary restraining order were no longer together at follow-up. The length of the relationship between the protected and restrained persons ranged from 1.5 to 24 years with a median and mean relationship length of 6 and 7 years, respectively.
Demographic Characteristics of Protected and Restrained Persons (n = 17).
Includes individuals on disability and those working “on and off.”
Relationship Characteristics of Protected and Restrained Persons (n = 17).
Questions about children were only asked in reference to the 6 months following restraining order issuance.
Abuse
Fifteen of the 17 study participants reported being the victims of various types of non–gun-related physical abuse (or serious threats) during the 6 months prior to obtaining the index temporary restraining order, most commonly being pushed, shoved, or grabbed. Ten of these women (66.7%) reported multiple types of abuse. More important, all of the women reported that the physical abuse had ended in the 6 months following the restraining order, but 5 women made one or more calls to the police to report threats of violence.
During the 6 months leading up to the index restraining order, 3 women reported being the victims of abuse or threats of abuse with a gun. One victim, who had been in a 3-year relationship with her abuser prior to getting the restraining order, reported that he held a gun to her head on two separate occasions and shot her once in the leg. Another victim said that her abuser had waved his gun at her in a threatening manner. The third victim reported that her abuser threatened to shoot her dog. Ten of the 17 women said they felt afraid of the restrained person during the 6 months preceding the restraining order because they knew he had access to a gun; 7 of these women said they continued to feel afraid even after getting the restraining order. None of the women in the sample reported gun-related abuse in the 6 months following the issuance of the restraining order.
Victims’ Knowledge of Restrained Persons’ Firearms
Regardless of how the restrained persons’ guns were discovered by law enforcement, all 17 of the study participants said that they were aware that the restrained person possessed one or more guns. Eleven women provided a precise number of guns belonging to the restrained person, and the remaining 6 women provided an estimate. The number of guns reported by the 11 women who provided an exact number ranged from 1 to 27, with a mean of 6 guns and a median of 2 guns. All of the participants reported that at least one of the restrained person’s guns was a handgun. When asked to indicate all of the ways in which they knew about the restrained person’s guns, 9 women said they had seen the guns, 8 said the restrained person showed his guns to her, and 4 said the restrained person told her he had guns. Less common responses included knowing about the restrained person’s guns because he belonged to a street gang, he hunted, she had gone shooting with him, and he gave her a gun for self-protection. In addition, two women reported that the restrained person held a job that required him to be armed; one was a security guard and the other was a highway patrol officer.
In comparing the survey responses with data from the Sheriff’s Office, we found that 5 victims reported that the restrained person was in possession of the same number of firearms as were accounted for by law enforcement, 3 reported that the restrained person had fewer guns than were accounted for by law enforcement, and 8 said that the restrained person had more guns than were accounted for by law enforcement. One notable example of the latter is a case in which the victim reported that the restrained person had 27 guns in his possession, but law enforcement only identified and removed 14 guns.
Firearm Removal From the Restrained Person
Eleven women reported that, when obtaining their temporary restraining order, either they (n = 6) or their attorneys or legal advocates (n = 5) asked law enforcement to take the restrained person’s guns or to make sure that he surrendered them (Table 3). All 5 of the women who did not request that law enforcement confiscate the restrained person’s guns said that they did want the guns to be relinquished or removed. Reasons for not asking included feeling afraid that their abuser would retaliate (n = 2), not being “absolutely certain” that the restrained person had a gun (n = 1), and wanting the restrained person to be able to keep his job as an armed security guard (n = 1). (One woman did not answer this question.)
Legal and Law Enforcement Responses to Firearm Removal Policy (n = 17).
Four of the 13 participants who saw a judge when applying for the temporary restraining order reported that the judge asked them whether the restrained person had any guns. Judges often issue temporary restraining orders based on written applications in the absence of more formal court hearings. Thus, all of the women were asked whether the judge directed law enforcement to take their abusers’ guns as the judge might have issued directions in person or in writing. Five women said the judge did direct the police or Sheriff’s Department to take his guns, 7 said the judge did not report such action, and 5 were unsure.
Eleven women reported that, after getting the temporary restraining order, a law enforcement officer asked whether the restrained person had any guns. Of these 11 women, 9 said that the officer asked whether they wanted the guns removed from the restrained person. All 9 women reported telling the officer that they wanted the guns removed.
In total, 8 of the 17 respondents reported that all of the restrained person’s guns were relinquished or removed (Table 4). Six reported that the restrained person kept some or all of his guns, and 3 reported being uncertain about whether the guns were surrendered or removed. Interestingly, in 4 of the 10 cases in which some or all of the restrained person’s guns were turned over to law enforcement, the protected person was responsible for physically giving the guns to law enforcement.
Fate of Restrained Persons’ Guns (n = 17).
Effect of Firearm Removal on Victims’ Feelings of Safety
Six of the 8 women whose abusers surrendered all of their guns reported that having the guns removed or surrendered made them feel safer; the other 2 said that having the guns removed did not affect how safe they felt (Table 5). Five of the 6 participants with partners who still had all or some of their guns reported feeling less safe; the other said that she did not feel more or less safe because her abuser’s guns were not turned in. Of the remaining 3 women who were uncertain about whether the restraining person continued to possess guns, 1 reported that not knowing made her feel less safe, and 2 said their uncertainty about whether he had guns did not make them feel any more or less safe.
Victims’ Feelings of Safety (n = 17).
Victims’ Pursuit of a Long-Term Restraining Order
Ten women reported that they obtained a final restraining order after their index temporary restraining order expired, one applied for but was not granted a final order, and 6 did not pursue a final restraining order. Reasons for not applying for a final restraining order were varied and included reconciliation with the restrained person, being physically unable to attend the required court hearing (for reasons not related to the abuse), not wanting to further disrupt the restrained person’s life, and being discouraged from going forward by an attorney. In addition, one woman reported that the restrained person’s attorney requested that the judge issue an “off the records” restraining order that would not be entered into California’s restraining order database so that it wouldn’t show on his record, even though this practice is against the law and negates potential protections such as abusers’ surrender of firearms. Two women reported that their abusers requested to get their guns back when the temporary restraining order expired; one of the 2 reported that he did, indeed, have his guns returned to him.
Discussion
Most of the 17 restraining order recipients surveyed wanted firearms removed from their abusers, and most of the victims for which this was fully accomplished felt safer as a result. Furthermore, 5 of the 6 women with batterers who continued to have access to firearms after obtaining a restraining order reported feeling less safe as a result.
The cooperation of judges presiding over domestic violence restraining order cases can be a vital tool in the effective removal of firearms from restrained persons (Frattaroli et al., 2011). Most of the women in this survey, however, reported that judges did not inquire about firearms or order their surrender. This finding is consistent with prior research in California and other states showing that judges often fail to order firearm relinquishment when issuing restraining orders or victims are often unaware of such orders if they have been made (Diviney, Parekh, & Olson, 2009; Moracco et al., 2006; Sorenson & Shen, 2005; Webster et al., 2010). In addition, study participants reported that the judge—at the request of the restrained person’s attorney—ordered that the restraining order not be entered into the state restraining order database, thus making the order’s firearm provisions unenforceable. A 2005 county-level study commissioned by the California Attorney General documented cases in which judges who disagreed with the firearm prohibitions on restraining orders crossed out these mandatory provisions. Several months after findings from the study were made public, many counties had greatly reduced this practice (Attorney General’s Task Force on Local Criminal Justice Responses to Domestic Violence, 2005; Seave, 2006). It appears that more work needs to be done to ensure that all judges are following the law.
Even when acting in accordance with the law, law enforcement officers did not always have the legal authority to remove all firearms from restrained persons. In cases where a victim informed the police that the restrained person was in possession of firearms but the restrained person denied their existence or refused to relinquish them, law enforcement officers often did not have the legal authority to search for and remove the firearms. Legislation authorizing the issuance of a search warrant to law enforcement when the property to be seized includes a firearm went into effect on January 1, 2010 (A.B. 789, 2009) and may prove useful in future law enforcement efforts to remove guns from batterers.
In our sample, it was common for abused women to believe that the restrained person was in possession of more guns than were accounted for by law enforcement. Although we cannot know the actual number of guns in the restrained person’s possession, one reason for this discrepancy may be the difficulty faced by law enforcement in identifying long guns. Under existing California law, records of all legal handgun transfers must be reported by all licensed firearm dealers so that they can be recorded in the Automated Firearm System (AFS), a centralized database maintained by the California Department of Justice. Legislation has been introduced in the California State Assembly during the past two sessions that would require the same reporting and record retention provisions for all long gun transfers as are currently in place for handguns (A.B. 18010, 2010; A.B. 809, 2011). The bill is likely to be reintroduced and, if passed, it will help law enforcement to identify long gun possession by individuals who are prohibited, such as persons under a domestic violence restraining order. The discrepancy between the number of firearms reported by victims and the number accounted for by law enforcement is likely due to abusers acquiring guns through transactions with private sellers, which are more difficult to regulate than is the case for licensed firearm retailers.
Some women who obtain temporary DVROs may be motivated to get long-term DVROs in order to obtain or maintain firearm prohibitions for their abusers. In this study, 10 of the 17 (59%) restraining order recipients obtained a final or long-term restraining order after the temporary order expired. This percentage of obtaining long-term DVROs is within the range observed in prior studies (37%-76%; Gondolf, McWilliams, Hart, & Stuehling, 1994; Vittes & Sorenson, 2006; Zoellner et al., 2000).
This study has several limitations. First, these data are based on a small sample of women from one urban county in California. Findings from the interviews cannot be generalized to other victims of intimate partner violence or to other settings (e.g., rural areas, counties in other states). Second, like most studies of domestic violence victims and other unstable and often transient populations, this study had a low response rate and a high noncontact rate. The women who were reached and agreed to participate may be different from those who were not reached or refused to participate. Third, these data are subject to recall and other biases associated with self-report data. Nonetheless, these interviews of victims whose armed abusers were targeted by a program to remove firearms from legally proscribed IPV offenders are the first of their kind.
In addition to surveys of women in other locales, future individual-level studies are needed to examine the effect of firearm removal on women’s actual physical safety. Vigdor and Mercy (2003, 2006) conducted two ecological studies looking at the effect of changes in state-level domestic violence firearm laws on homicide rates. They found that laws keeping firearms from persons under a restraining order reduce female IPH victimization rates, but only in states with a system to screen potential firearms purchasers for restraining orders. The studies found no reduction in IPH rates associated with laws intended to keep guns from domestic violence misdemeanants or with laws allowing firearm confiscation by law enforcement at the scene of a domestic violence incident. As noted by the authors, these findings may be explained by weak or incomplete implementation of the laws.
These studies are valuable in establishing a connection between firearm relinquishment laws and related homicides, but they could not measure implementation and enforcement. Future individual-level data could be used to establish a connection between implementation and enforcement of these laws and firearm-related assaults by intimate partners.
Most restraining order recipients clearly wanted firearms removed from their abusers even if the physical violence had ended. These women’s desires are in concurrence with state and federal law making access to firearms illegal for persons under a domestic violence restraining order. This program was one of the few to actively enforce these laws and can be used to inform future such efforts.
Footnotes
Acknowledgements
The authors gratefully acknowledge Vanessa McHenry for her assistance with the administration of the survey and the interviewers for fielding the questionnaire. We also thank the California Department of Justice, Bureau of Firearms, and the San Mateo County and Butte County Sheriff’s Offices for their support of and important contributions to this project.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the California Department of Justice (Agreement Numbers 06-7816 and 10-7816).