Abstract
Research on covictims, family members, and close friends who have lost loved ones to intimate partner homicide (IPH) is a neglected area of study. We conducted phenomenological interviews with covictims to gain insights into risk and lethality, examined affidavits from criminal case files, and reviewed news releases. The data uncovered acute risk factors prior to the homicide, identified changes in the perpetrators’ behavior and the perpetrators’ perceived loss of control over the victim, and described barriers that victims faced when attempting to gain safety. Findings suggest that recognizing acute risk factors is an important area for future IPH research.
Research estimates that between 4.8 and 5.3 million episodes of intimate partner violence (IPV) occur each year in the United States (Centers for Disease Control and Prevention [CDC], 2011; Klap, Tang, Wells, Starks, & Rodriguez, 2007). These incidents may include physical and sexual violence, threats, and emotional abuse (McLeod, Hays, & Chang, 2010; Tjaden & Thoennes, 2000). The CDC (2010) cites IPV as a preventable public health problem and a key health issue in the United States today. To add to the alarming facts about the episodes of IPV, data from the Bureau of Justice Statistics (2012) indicate that intimate partners committed 14% of all homicides in the United States in 2007.
Approximately 40 years of IPV research, drawn from the knowledge and experience of victims, advocates, practitioners, and researchers, has been used to guide legislation, create protocols for first responders, alter policies and practices, and enhance the well-being of victims. One important area of continued expansion of knowledge focuses on risk and lethality for intimate partner homicide (IPH). This growth has coalesced around a number of areas, including factors that contribute to the greatest risk of victimization (Aldridge & Browne, 2003; Campbell, Glass, Sharps, Laughon, & Bloom, 2007; Garcia, Soria, & Hurwitz, 2007), risk factors for reassault, and risk factors that increase lethality (Campbell, 2004).
A burgeoning area of research focuses on the identification of key risk factors for IPH, an area of research far less studied than IPV and more difficult to predict (Aldridge & Browne, 2003; Campbell et al., 2007; Campbell, Webster, & Glass, 2009; Campbell et al., 2003b; Garcia et al., 2007). Methods to increase knowledge about risk of IPH have included research with advocates and analysis of medical examiner reports and police records (Campbell et al., 2003a). More recently, researchers have enlisted family members and close friends of IPH victims to increase understanding of IPH risk factors (Glass, Laughon, Rutto, Bevacqua, & Campbell, 2008). We see this line of inquiry as promising because family members and close friends are often the source of firsthand and intimate knowledge regarding the relationship between the victim and the perpetrator. We define these family members and close friends as covictims because of their loss of loved ones to a homicide. However, incorporating the knowledge and experiences of covictims remains an underutilized area of research, despite the potential for increasing our understanding about the events leading up to IPH that covictims often possess. Indeed, we located only three studies in which the knowledge and experience of covictims of IPH were examined (Campbell et al., 2003b; Hardesty, Campbell, McFarlane, & Lewandowski, 2008; McNamara, 2008). The current exploratory study adds to this growing body of research on risk factors for IPH (Campbell et al., 2007; Campbell et al., 2009; Campbell et al., 2003a, 2003b; Farr, 2002). Before describing the study, we first discuss the path research has taken in IPV and IPH.
IPH
Homicide trends in the United States indicate that violent crimes, specifically homicides, have declined throughout recent decades; however, the rate of decline for female victims of IPH is approximately one half the rate of male victims (Fox & Zawitz, 2007). Moreover, the FBI reports that the rate of homicide of women who were killed by husbands (including common law and ex-spouses) or boyfriends (including ex-boyfriends) increased from 32.9% in 2007 to 37.5% in 2010 (U.S. Department of Justice–FBI, n.d.). During 2010, women made up nearly 82% of victims killed by an intimate partner (U.S. Department of Justice–FBI, n.d.). More recently, an examination of all homicides–suicides in the United States between January 1 and June 30, 2011, revealed that 72% of all cases involved intimate partners, and 94% of the victims were women (Violence Policy Center, 2012). Overall, these statistics and national research studies bear out that, compared with men, women are at greater risk of being victims of IPV and IPH (Campbell et al., 2007; Tjaden & Thoennes, 2000).
A feminist approach to understanding and explaining this enormous disparity between the death rate of women and men as victims of IPH has been utilized for years by a number of researchers and practitioners. A recent thorough review and discussion focusing on the policy implications of the feminist perspective on femicides demonstrates the efficacy of a feminist perspective compared with other frameworks (Taylor & Jasinski, 2011). As Taylor and Jasinski (2011) and others have noted, feminist analyses of the causes of IPV support a broad structural perspective that examines the ways in which larger community and societal issues, such as gender inequality and male social control and entitlement, permeate the foundation of attitudes that condone violence against women, including one of its lethal forms, IPH (see, for example, Bograd, 1988; Echeburúa, Fernández-Montalvo, de Corral, & López-Goñi, 2009; Ellis & DeKeseredy, 1997; Hunnicutt, 2009; Taylor & Jasinski, 2011). A feminist perspective is the one that we believe is the most efficacious and informative for comprehending the disparity between the percentage of murdered women who are killed by their intimate partners compared with the percentage of men in that category. Furthermore, understanding the risk factors within these relationships will aid in the recognition of risk and lethality.
Risk Factors for IPH
Researchers agree that previous abuse in the relationship is one of the most important risk factors of IPH (Campbell et al., 2007; Garcia et al., 2007). Gallup-Black (2004) recognized IPH as a culminating event following a pattern of abuse. This pattern of abuse is difficult for the victim to terminate because separation, or even threat of separation, is another risk factor of IPH (Campbell et al., 2003b; Garcia et al., 2007). Research has also shown that any threat of leaving the abuser is dangerous because the abuser loses a sense of control over the victim. Indeed, Campbell and colleagues (2003b) reported that women who were separated after living with the perpetrator were over 3 times more likely to be victims of homicide than those who had not been separated. In addition, Garcia and colleagues (2007) found that the likelihood of homicide increased even more if the victim left the abuser for another partner. For women who left their partners and were later killed, the homicide usually occurred within the first year following separation, indicating that both separation and previous abuse are predictive risk factors for the next 12 months (Campbell et al., 2007).
Stalking is another widely recognized risk factor of IPH. In a review of spousal homicide, Aldridge and Browne (2003) reported that 91% of victims of attempted homicide who had reported being abused also reported being stalked by their partners. Campbell and colleagues (2007) concluded that stalking behavior may be even more predictive of IPH than of IPV. Supportive of that finding, Aldridge and Browne (2003) reported that 23% of femicide victims experienced stalking prior to the event.
Researchers have noted the availability of handguns and other weapons as risk factors of IPH. Male perpetrators used handguns in roughly half of all IPH cases in the United States (Aldridge & Browne, 2003). The presence of a gun in the home can also be a means of threatening the victim. The proximity of lethal weapons is especially dangerous when coupled with alcohol or drug abuse by the abuser, another risk factor of IPH. Aldridge and Browne (2003) noted that 48% of the men had consumed alcohol at the time of the homicide. Another study revealed that men who drank alcohol were 3 times more likely to be violent toward their partners than those who did not drink alcohol (Garcia et al., 2007). Furthermore, Campbell and colleagues (2007) noted that although alcohol use is not recognized to be as strong a predictor as other factors, 70% of male perpetrators in one study reported using drugs and/or alcohol at the time the homicide occurred.
Overall, researchers have found that previous abuse, separation or threat of separation, stalking, the availability of handguns, and alcohol or drug abuse are risk factors of IPH (Aldridge & Browne, 2003; Campbell et al., 2007; Campbell et al., 2003b; Garcia et al., 2007). Unfortunately, many of the risk factors for IPH are similar, if not the same as those for IPV, thus making predictions of IPH more complex. Clearly, not all victims of IPV are murdered by the perpetrator of the violence. But what makes some cases end in death, and what are the risk factors for IPH? A step in unraveling types of risk is to focus on the experiences of covictims of homicide who had firsthand knowledge of the relationship between the victim and the perpetrator. Qualitative inquiry provides the method for exploring a little known subject with individuals who have experienced the phenomenon.
The Current Study
The current exploratory research, part of a larger study in one small, rural Northeastern state, focused on lessons learned from covictims regarding risk and lethality. The study was a collaborative endeavor by university researchers, the state’s Office of Victim/Witness Assistance (OVWA), and the Federal Office for Victims of Crime. Three university researchers in conjunction with OVWA created an interview guide to be used with covictims. In addition to questions about needs and services, we sought covictims’ insights into and knowledge about events leading up to their loved ones’ homicides that might be informative to practitioners and researchers.
Method
Participants
We assigned pseudonyms to the seven male and seven female covictims who comprised the sample. All participants self-identified as Caucasian, and English was their primary language (N = 100%), which mirrors, to a great extent, the composition of the citizens of the state (93.9% Caucasian; U.S. Census Bureau, 2012). Participants included mothers (n = 2), fathers (n = 4), stepmothers (n = 1), stepfathers (n = 1), adult children (n = 1), aunts/uncles (n = 1), sisters (n = 1), and close friends (n = 3). Participants’ ages ranged from 37 to 77 years. The mean length of time since the homicide was approximately 6 years. Not interviewed, due to our concern regarding potential retraumatization, were an additional 16 covictims under the age of 18 at the time of the homicide, who were the children of the victim or the perpetrator.
Procedure
We used a multi-method approach, known as triangulation (Fontana & Frey, 2005), to understand risk and lethality of IPH relationships. Thus, we conducted interviews with covictims, reviewed news releases, and examined affidavits from each of the cases. Some news releases were posted on the state attorney general’s website, whereas others were published in statewide and local newspapers. Affidavits are sworn statements made by police officials, which include the facts of the case and are part of the case record. The news releases and affidavits were sought on completion of the interviews and served as additional sources of data.
University researchers collaborated with advocates in the OVWA in the state attorney general’s office to determine eligibility for inclusion in the study. OVWA advocates provide covictims with information about the criminal justice process and also offer emotional support from the time of the homicide through post conviction, and thus, possess a depth of knowledge about all of the IPHs in the state. Advocates identified potential participants/covictims from all IPHs between 2000 and 2009. On approval of the study from the university’s institutional review board (IRB) for the protection of human participants, advocates made the initial contact with covictims utilizing a telephone script created by the researchers. Covictims who consented to participate were mailed follow-up letters to confirm their initial agreement and inform them that a researcher from the university would contact them by telephone. This letter was followed by a call from one of the researchers to schedule interview meeting places and dates. If no contact could be made after a second message, the participant’s name was removed from the list.
Data Collection
Advocates identified 59 covictims from 31 IPHs. Contacting potential interviewees proved to be difficult as many had either relocated without leaving contact information or changed their telephone number. Of the 59 possible interviewees, 16 covictims, representing 9 homicides, agreed to be interviewed. Two of the 4 covictims from 1 case chose to withdraw from the study, resulting in a final sample of 14 covictims from 9 cases. Interviews were conducted over 8 months between June 2010 and February 2011.
Given the sensitive nature of the subject, we chose to use semi-structured, face-to-face, phenomenological interviews, which provided flexibility for the researchers and interviewees. This interview format enabled researchers to obtain detailed descriptions of IPV and other aspects of the victim–perpetrator relationship. Moreover, semi-structured interviews allowed for a more conversation-like atmosphere, conducive to sharing sensitive narratives. Although the interviews were largely free-flowing, specific topic areas included the nature of the relationship between victim and perpetrator, case characteristics, service utilization patterns by the victim, and the covictims’ perspective on level of dangerousness. Covictims in two of the cases requested telephone interviews. In those interviews, we secured permission to use the speakerphone feature of the telephone and followed the same format as with face-to-face interviews (e.g., the principal investigator conducted the interview while the second researcher took extensive notes). We chose not to audiotape the interviews so as not to further upset the covictims by using a method that might emulate those used by the police, prosecutors, and media. Rather, we chose to create a less formal atmosphere to decrease covictims’ discomfort in telling the story of their loved one’s murder. The note taker was free to ask clarifying questions should the need arise, further emphasizing a conversation-like atmosphere. Following the interview, notes were reviewed with the covictim to ensure accuracy. This method of data collection proved invaluable in creating an atmosphere of attentiveness on the part of the researcher and allowed for expression of emotion from the covictim without the presence of a recording device. In addition, reviewing the interview with the covictim immediately after its conclusion made certain that the interviewers captured what the covictim had conveyed.
The cases reported in the current study represent one third of the IPHs in the state during the time period under review. Interviews were conducted at a site chosen by the interviewee, including covictims’ homes and local libraries. A total of 9 interviews with 14 participants formed the data for this study. Each interview lasted 2 to 3 hr. During that time, the researchers indicated to interviewees that they were free to take as many breaks as they wanted. We discussed confidentiality, reviewed the IRB-approved letter of informed consent, and explained that they did not have to answer any questions they did not wish to answer and could withdraw from the research process at any time without penalty. Participants were given a list of resources should they wish to talk with someone for emotional support. As a token of our appreciation, we offered participants a US$25 gift card to one of three retail stores.
Heideggerian Hermeneutic Phenomenology
We followed a seven-stage process of analysis to carry out a hermeneutic interpretation of the interview data. This approach is grounded in Heideggerian hermeneutic phenomenology and was used to explicate the lived experience of the 14 covictims (for a complete review of this interpretivist methodology, see Diekelmann, Allen, & Tanner, 1989; for an example of this methodology utilized to interview sexual assault survivors, another group of traumatized narrators, see Murphy, Moynihan, & Banyard, 2009). This philosophy and methodology are concerned with the meanings that individuals make of their experiences, acknowledge that meanings are embedded within a particular historical and cultural context, and allow both researchers’ and participants’ openness to different interpretations. In the study, Heideggerian hermeneutic phenomenology provided a framework for reflecting, interpreting, and gaining insights into the meaning that the covictims attributed to their experiences. The covictims’ words became the texts for analysis. According to Heidegger (1953/1996), phenomenology should make manifest what is hidden in ordinary everyday experiences, which makes it particularly appropriate as an application to this study. We chose this form of analysis to elicit covictims’ experiences and knowledge of the events leading up to the homicide.
The constant process of moving from the text in its entirety (e.g., considering all nine transcripts as one text) to its parts (e.g., considering each transcript as a separate entity) and back again forms the process of whole–part–whole, or the hermeneutic circle (Benner, 1994). Ultimately, we uncovered the three themes (Changes in Behavior; Loss of Control; and Barriers to Help) that we discuss in the “Findings” section below.
Data Analysis
There were 16 deaths (9 men and 7 women) that included 9 IPH victims, 2 close friends of the victims, and 5 perpetrators who took their own life. In addition, three people (two men and one boy) sustained life-threatening injuries as a result of attempted homicides. Of the nine perpetrators, seven were men. Comparatively, 86% of all the IPH perpetrators in the state between 2005 and 2007 were men (New Hampshire Coalition). Consistent with prior research (Campbell et al., 2007; Garcia et al., 2007), a history of domestic violence was present in each of the nine cases. Eight of the nine homicides occurred in a home that the victim and perpetrator once shared. The weapons used by the perpetrators in the homicides included six guns (all used by men), one knife (used by a woman), and multiple weapons (used by one man and one woman). In one case, assistance had been sought from health care professionals in the 48 hr prior to the homicide. One victim sought assistance from a domestic violence advocate prior to the homicide. Six victims had been in contact with law enforcement in the months prior to the homicide, and four of those victims had contacted the police within 48 hr prior to the homicide.
Findings
Changes in Behavior
Changes in the perpetrator’s behavior were noted in seven of the nine cases, signaling an increase in risk to the victim. Jenny, best friend of Kathy who was murdered by her husband, told us about changes in the perpetrator’s behavior during the weeks just prior to the homicide:
The month before Kathy died, there was an increase in arguments. Her husband’s drug use increased . . . He wasn’t afraid to hide it near the end . . . The night of the murder she told him that their marriage was over . . . He took the axe to her and [also] shot her . . . [I think it was a] combination of control—his knowledge that the marriage was over and the crack cocaine.
The couple had frequent arguments over the husband’s drug usage. During some of those arguments, the perpetrator became violent, and Kathy obtained a restraining order on at least one occasion. In the final weeks leading up to the homicide, the perpetrator agreed to attend counseling, and Kathy subsequently dropped the restraining order. In a few short weeks, the perpetrator’s drug use increased again and, in response, Kathy declared that their relationship was over. The combination of the perpetrator’s increasing drug usage and Kathy’s declaration about the end of the marriage culminated in a particularly brutal murder while their three children stood outside the locked bedroom door.
Other covictims noted that the victims took action in response to changes in the perpetrator’s behavior. Sometimes, victims’ actions included talking to others about their fear. Sherry, best friend of Jean, told us that Jean’s estranged spouse had recently become ill, and, because of his illness, he had agreed to give Jean temporary custody of their 11-year-old son. Jean became increasingly concerned about how her estranged spouse might respond to the change in custody and expressed her concern to the police. Sherry remarked that Jean’s fear was all too real:
He [perpetrator] showed up at the house and said he would give her money to buy their child underwear because she didn’t have any . . . Jean stepped out the door [to take the money] and he shot her while their son was inside the house.
For some perpetrators, an impending divorce hearing or being rebuffed by another woman could trigger a fatal change in their behavior. Guy, whose daughter Darlene was killed by her husband, reported that as the divorce hearing approached, his son-in-law’s behavior changed to include stalking. The victim became increasingly alarmed about the change in behavior and contacted her estranged spouse’s former wives who confirmed his abusive pattern during their marriages as well. Guy remarked, “Yes, I was worried before she threw him out. He is an ex-cop that walks around with a pistol in his briefcase.” In response to her mounting fear, Guy gave his daughter a rifle and ammunition to protect herself. With growing fear, Darlene obtained an order of protection just days prior to her murder. The night of the homicide, which was the estranged couple’s first wedding anniversary, the perpetrator had a date with another woman who rebuffed his attempts at intimacy. A few hours later, the perpetrator drove to the home he had shared with Darlene, entered the house through the front door, and shot her in the back as she stood in front of the clothes washer.
Although changes in the perpetrators’ behavior just prior to the homicide was the most common story told to us, other covictims, talked about changes in the victim. Patti, daughter of Ginny who was murdered by her husband, told us that her mother informed her husband on numerous occasions that she wanted a divorce, but he refused to move out. Finally, in desperation, Ginny moved 1,000 miles away from him only to be stalked. He found and moved back in with her against her wishes. Within 1 year, the perpetrator shot and killed her and took his own life. We asked Patti to describe the type of relationship that her mother had with the perpetrator and she said, “almost a hunt and prey type.” Patti talked about how her mother’s relationship with her spouse weighed greatly on Ginny:
[It] sounded like mom was ready to give up. She would say, “The good Lord takes care of us.” She sounded tired. No matter what we said, what we did . . . She was tired.
Similarly, Steven and Gloria told us a story about their son, Kevin, whose behavior changed in response to his girlfriend. Kevin met his girlfriend/perpetrator at a drug rehabilitation facility where they were both in recovery. The couple lived together for 10 months after leaving the facility and had an ongoing struggle maintaining sobriety. Despite these problems, the couple was employed and, according to Gloria, seemed to be doing well. However, a few weeks prior to his murder, Kevin believed his girlfriend was using drugs again and was having an affair. Gloria told us,
It seemed like they were getting their act together. I thought he had more of a purpose to get off heroin . . . The day before he died, we went to the mall to buy new clothes. Then we saw a kitten he wanted to get for her [his girlfriend/perpetrator] . . . He said the next day he was getting off early from work because the two of them didn’t get to see each other much because of their work schedules.
Despite the appearance that things were going well in the relationship, Kevin’s fear that his girlfriend was having an affair led him to inquire with friends and acquaintances about her fidelity. On the night of the homicide, he questioned his girlfriend and a violent episode ensued. Violence escalated and ended with the perpetrator taking Kevin’s life. Previous research has found that the inclusion of drugs and alcohol in the lives of a couple can have a profound impact on the relationship and may increase violence. Particularly in this relationship, an additional risk factor seems to have contributed to this tragic ending.
Loss of Control Over the Victim
Loss of control was evident in eight of the nine cases. Covictims offered insights as they explained the perpetrator’s attempts to maintain power and control over the victim. Jim told us that when Angela’s husband’s violence escalated, she sought support and counsel from a male neighbor whom she had known during her high school years. The perpetrator became suspicious of Angela’s relationship with their neighbor and searched Angela’s phone for evidence of conversations between them. Violence escalated as the perpetrator sensed Angela was pulling away from him, resulting in an increase in verbal threats, physical violence and sexual assault. Forty-eight hours prior to her death, Angela’s husband attempted to strangle her, and beat her so severely that she sought medical treatment, spoke with a crisis counselor, reported the crime to police, and obtained a restraining order. Her husband was arrested for strangulation, but was released from jail the same night. Jim told us the following story:
Things definitely escalated toward the end. I had no idea he [perpetrator] was out of work. There was a lot of yelling and screaming between them and then one night he held a pistol to her head. His jealousy escalated as well. She [Angela] was communicating with a male friend from next door. She was reaching out for help . . . telling her friend about the abuse. Her husband got a hold of her cell phone and figured there was a relationship between them . . . In the last 5 weeks, my daughter told me to get a room ready for her and their son. She stayed for just one day. I knew the jealousy issues and things like that. I know her husband broke down a door in their house and that was about one week before he killed her.
Forty-eight hours following the strangulation and receiving the notice of the restraining order, the perpetrator learned that Angela and Jim were going to the house to retrieve some of Angela’s personal belongings. The perpetrator entered the house armed with several guns, awaited their arrival, shot and killed Angela, critically wounded Jim, and took his own life.
Another example of loss of control was told to us by Linda, sister of John. John and his girlfriend were both murdered by her estranged spouse the night before the final divorce hearing. The perpetrator and his estranged spouse had been separated for 2 years and had multiple divorce hearings scheduled. Each time a hearing was scheduled, the perpetrator, with the assistance of his attorney, used legal tactics to delay the court process. Two days prior to the double homicide, the perpetrator went to the home he had once shared with his wife and their two children. He verbally threatened his estranged wife and physically threatened John. Linda described her concern for her brother’s safety the day prior to the murder: “When we were told he [perpetrator] canceled the lawyer, something just didn’t add up.” The night of the homicide, the perpetrator entered his estranged wife’s home while she and John and the two children were asleep. He shot and killed his wife and John, and then killed himself. It seems that the fact of his estranged wife living with another man may have symbolized the perpetrator’s loss of control over her. Canceling the meeting with his attorney might be viewed as the perpetrator’s resignation to divorce, or it might be seen as the final steps to completing a plan to take the lives of his estranged wife and her boyfriend.
Sherry’s narrative about her friend Jean’s relationship with her perpetrator reminds us that loss of child custody also signals loss of control in the relationship. Within 48 hr of a change in custody, the perpetrator killed the victim, critically injured their child and Sherry’s husband, terrorized Sherry’s children, and took his own life. Other cases revealed similar evidence of the perpetrator losing control through some form of estrangement, such as a geographic move as we heard from Patti, or the murder of Kathy within hours of telling her husband that their relationship had ended.
Barriers to Help
Covictims in the current study talked about different types of barriers that the victims faced when they sought help prior to their murder. In some cases, the perpetrator created the barriers as part of the cycle of abuse, whereas in other cases, victims faced systemic barriers related to organizations. Each covictim spoke of increasing risk to the victim and each talked about her or his concern for their loved one. Many talked about steps that the victims had taken to be safe, such as reaching out to domestic violence shelters, seeking help from loved ones, leaving, moving, creating geographic distance, using the justice system, or telling others about the abuse. Most of the victims in the current study were in the process of leaving the relationship, and the barriers they faced may be explained by examining systems that did not work as well as they could.
Patti told us that her mother, Ginny, had asked her husband for a divorce several times. When her husband resisted, she moved across the country, which resulted in being stalked and found in the remote part of the state where Ginny was hiding. Patti remarked that her mother had said, “I can’t get rid of him. It’s just that. When I’m with him, I can watch him.” Eventually, her husband moved back into the home despite her pleas. Ginny’s sense of hopelessness was palpable. Barriers in this case appear to have stemmed from the geographic isolation of the area as well as the barriers that the perpetrator created by isolating Ginny from her adult children and grandchildren, her church, and all support systems.
Some of our interviewees discussed barriers emanating from large systems, such as the justice system, mental health systems, and the community at large. In three cases, covictims expressed frustration that the police did not do all they could have. Four victims either had current or recent orders of protection or restraining orders against their perpetrators prior to the homicide. Guy’s comment (father of a victim) summarizes the sentiments of many:
They pass a lot of laws, but they can’t enforce them. How are you going to stop someone? A piece of paper in her hand [restraining order] isn’t going to stop anyone.
Jim vented his frustration when a previous domestic violence incident did not result in harsher punishment for his daughter’s perpetrator: “I’m surprised [about] the strangulation and throwing someone down the stairs . . . To me, that’s attempted murder, and he was in and out of jail the same night.” Jim indicated that his daughter had taken all the appropriate steps to reach out for protection by contacting all the systems in place to protect victims. Despite all of her precautions, her husband killed her within 48 hr of the strangulation. Gerald and Shirley, parents of Alan who was murdered by his girlfriend, noted that the police did not actively intercede on behalf of their son even after seeing evidence of physical injury. Forty-eight hours later, Alan was murdered. Covictims in two other cases revealed that despite procurement of restraining orders and assurances by police that their houses would be watched, when their loved one reached for protection, the safety net was not there.
Some covictims revealed frustrations with mental health and substance abuse treatment systems and pointed to the barriers to getting help. Stephen and Gloria, whose son was murdered by his girlfriend, stressed the importance of establishing a statewide educational service to inform citizens of the addictiveness of certain drugs. From Stephen’s perspective, lack of access to substance abuse services was a major barrier to his son’s ability to recover from addiction and participate in a healthy intimate relationship.
Constitutive Pattern
The themes in the current study elucidate the constitutive pattern, Hanging in the balance: The precarious nature of safety. Abuse and violence had become part of the everyday lives of the victims in our study. The constitutive pattern reflects the delicate balance in which the victims attempted to maintain a stable life and also find safety, sometimes for themselves and often times for others. The precarious nature of safety was severely challenged by the presence of a triggering event, such as a change in child custody or the victim’s decision to end the relationship. The presence of these events signals to the perpetrator that he or she has lost control over the victim, and becomes the impetus for a dramatic change in behavior, which we define as an acute risk factor, ultimately leading to homicide.
An example can best be seen in the story of Jean, told by Sherry. Jean had been awarded temporary custody of her son due to the perpetrator’s ongoing health issues. The change in custody became the triggering event, which led to the perpetrator’s increased loss of control and increased violence. “It was a temporary [custody] order I think when [the perpetrator] realized that it was real, that [their child] was going to stay with Jean. I think he freaked. I think he snapped.” The sudden increase in extreme violence in the form of homicide occurred 1 week after a triggering event in eight of the nine cases. Time seems to be linked inextricably with the onset of acute risk. Although covictims did not recognize the perpetrator’s sense of loss of control nor expected the relationship to end tragically, each case revealed a similar pattern of a triggering event, loss of control, and change in perpetrator’s behavior, which led to extreme violence. And in eight of the nine cases, the element of time played a significant role.
The cases in the present study reveal a clear picture of the relationship between victim and perpetrator as well as the victims’ final days. Covictims helped to clarify the presence of two types of risk factors in these relationships. The first set of risk factors include acts that are commonly recognized in the research literature and are referred to as static factors (Cattaneo & Goodman, 2005). The presence of multiple static factors has been found to increase lethality in relationships (Campbell et al., 2007; Garcia et al., 2007). We found multiple static factors in our study, such as isolation, drug use, and physical violence, and we distinguish them from a second set of risk factors, which we refer to as acute risk factors. Examples of acute risk factors include severe physical violence, a change in custody, and strangulation (see Table 1 for a complete list of static and acute risk factors in the current study).
Case Characteristics.
All the names are pseudonyms.
Discussion
Covictims described stories of entrapment, stalking, strangulation, and brutality; and in those stories, we heard the many ways victims sought to break free from the relationship. Victims sought assistance from friends and family, health care professionals, and the justice system. In five of the nine cases, the homicide was followed by the perpetrator taking his own life, the ultimate response to loss of control over one’s partner. Two of the three themes that emerged, Changes in Behavior and Loss of Control, refer to patterns that relate specifically to the perpetrators’ actions and thoughts. The third theme, Barriers to Help, refers both to barriers constructed and maintained by the perpetrator, and barriers created by larger systems. Taken together, the three themes brought to light the victims’ common experiences of violence and abuse in the last few days leading up to the homicide.
Prior research has acknowledged the overlap of risk factors for IPV reassault with risk factors for IPH (Campbell, 2004), but has also noted a group of factors that indicate a significant risk to victims that may culminate in homicide (Campbell et al., 2009). We concur with that finding, and we extend it to refer to events in the life of the couple, which may provoke a spike in the perpetrators’ realization of loss of control and ultimately lead to increasingly abusive acts. The constitutive pattern illuminated the presence of an acute risk factor, which was triggered by an event that challenged the perpetrator’s sense of control over the victim and relationship, and led to changes in the perpetrator’s behavior, culminating in homicide. We also learned from a covictim about an exception, Ginny, who was stalked and killed 1 year after leaving her husband and being found again. In this case, evidence of an immediate triggering event does not seem apparent. However, the pattern identified in other cases, such as the perpetrator’s perceived loss of control over the relationship and changes in his behavior, manifested in the story of Ginny as well. Although acute risk factors have not been a focus in IPV or IPH research, our findings suggest that this is an important area for consideration by professionals working with victims of IPV and a focal area for future research.
Limitations, Implications, and Directions for Future Research
We recognize that our exploratory study has a number of limitations and that our findings are most important in their suggestions of directions for future research. First, our findings were based on a small sample of covictims that occurred over a 10-year period in one small, rural Northeastern U.S. state. The state is relatively homogeneous in terms of racial composition and nonurban character and has a low homicide rate in comparison with other states. The FBI reports that only three states had lower murder rates in 2010 (U.S. Department of Justice–FBI, n.d.). Second, the issue of participants volunteering is always a consideration when reviewing the possible reasons a person might choose to participate in a study. Finally, without a comparison IPV group, we cannot know whether the risk factors identified in these cases differ from risk factors in relationships that do not end in homicide.
Although the current exploratory study is based on a small sample, much can be learned from the experiences of covictims. It is important to note that the covictims in our study did not recognize an increase in risk and lethality, and none of them thought that the relationship would result in murder. They spoke of erratic and extreme behaviors by the perpetrator that would have raised alarm as risk factors for professionals who work with victims of IPV. Covictims increased our understanding of the time period just prior to the homicide and helped elucidate acute risk factors. In addition, covictims have been helpful in clarifying the differences between survivors of abuse and femicide victims (Campbell et al., 2003b; Glass et al., 2008).
Future research with a larger heterogeneous sample of covictims would add to the burgeoning body of literature regarding risk and lethality and would assist in the creation of a typology of risk and lethality factors. Likewise, the construct of time is another area where more research on the interplay of acute risk with the temporal element related to the homicide is needed. Multiple opportunities exist for extending the current research, including a broader circle of covictims and moving beyond the confines of a small state with limited diversity. Extending this exploratory study would help bring to light the chain of events as risk factors shift from static to acute, fostering an improved evaluation of danger.
Overall, the current study extends prior research on risk and lethality in violent intimate relationships in a number of ways. By highlighting the contributions of an underutilized group, covictims of IPH, changes in behavior, and perceived loss of control on the part of the perpetrator were identified. The research also clarified the distinction between types of risk factors and located the presence of acute risk factors in the lives of eight of the nine victims, underscoring the prominence of time as a factor related to lethality. Finally, the constitutive pattern and the three themes—Changes in Behavior, Loss of Control, and Barriers to Help—elucidated the events leading up to homicide and brought to light the common experiences of the victims in our sample.
Footnotes
Acknowledgements
Acknowledgement goes to Abbie Beaulieu for her many contributions to this project.
Authors’ Note
The content is solely the responsibility of the authors and does not necessarily represent the official views of the U.S. Department of Justice, Office for Victims of Crime, or the Office of the New Hampshire Attorney General.
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 a grant from the U.S. Department of Justice, Office for Victims of Crime, Intensive Case Management Family Members of Homicide Victims in Rural Areas Project (ICMP), Office OVC Contract/Cooperative Agreement 2009-VF-GX-K008 under CFDA 16.582.